Research


Publications


  • In Press

    Huang, F., & Cornell, D. (in press). Teacher support for zero tolerance is associated with higher suspension rates and lower feelings of safety. School Psychology Review.

    Konold, T.R., & Sanders, E.A. (in press).  A latent approach to estimating rater effects and differences in multilevel contexts: A school climate illustration.  Measurement: Interdisciplinary Research and Perspectives, 18(4).

  • Publications 2020 to Present

    Cornell, D. (2020). Threat assessment as a school violence prevention strategy. Criminology & Public Policy, 1–18. https://doi.org/10.1111/1745-9133.12471

    This paper describes the immense difficulty of predicting that someone is going to carry out a school shooting and then turns to threat assessment as a more promising violence prevention strategy. In schools, a multidisciplinary threat assessment team investigates reported threats and develops responses that are calibrated to the seriousness of the threat and the student's educational needs. Researchers have found that school teams have been able to resolve thousands of student threats with no serious acts of violence, yet permitting the majority of students to return to school. Controlled studies have found that schools using this approach can have reductions in the use of school suspension and improvements in student and teacher perceptions of school climate.

    Threat assessment represents a fundamental shift in the risk assessment field away from the pursuit of predictive accuracy toward a broader approach to the prevention of violence by helping troubled individuals. Threat assessment offers schools a proactive alternative to reactive practices such as zero tolerance discipline and costly investment in building security measures.

    Crowley, B. Z., & Cornell, D. (2020). Associations of bullying and sexual harassment with student well-being indicators. Psychology of Violence. Advance online publication. http://dx.doi.org/10.1037/vio0000345

    Objective: Although both school bullying and sexual harassment have been widely studied, there is little research comparing their prevalence and impact within a single sample. The present study investigated three research questions: (a) What is the prevalence of bullying compared with sexual harassment in high school, and how does prevalence differ across gender, grade, race/ethnicity, and socioeconomic status? (b) How is bullying associated with student well-being indicators, as compared with sexual harassment? (c) What is the association of just a single experience of sexual harassment with student distress? Method: A statewide survey of 85,750 students (Grades 9–12) in 322 high schools reported how many times in the past school year they had experienced different types of bullying and sexual harassment. Participants also reported about mental health, risk behaviors, academic achievement, student engagement, and feelings of safety. Results: Sexual harassment was slightly more prevalent than bullying, but both demonstrated meaningful associations with student well-being indicators. Even a single experience of sexual harassment was associated with higher student distress, with experiences of sexual rumors being the most distressing. Conclusions: Researchers should be aware that bullying and sexual harassment are prevalent in schools and associated with negative well-being indicators. Attention to either alone would not yield an adequate assessment of adolescent victimization experiences, and prevention efforts should consider both forms of aggression to provide safer, healthier learning environments for students.

    Edwards, K. D., & Konold, T. R. (2020). Moderated mediation analysis: A review and application to school climate research. Practical Assessment, Research, and Evaluation, 25, Article 5. Available at: https://scholarworks.umass.edu/pare/vol25/iss1/5

    Moderated mediation analysis is a valuable technique for assessing whether an indirect effect is conditional on values of a moderating variable. We review the basis of moderation and mediation and their integration into a combined model of moderated mediation within a regression framework. Thereafter, an analytic and interpretive illustration of the technique is provided in the context of a substantive school climate research question. The illustration is based on a sample of 318 high schools that examines whether school-wide student engagement mediates the association between the prevalence of teasing and bullying (PTB) and academic achievement on a state-mandated reading exam; and whether this indirect effect was moderated by student perceptions of teacher support.

    Maeng, J., Malone, M., & Cornell, D. (2020). Student threats of violence against teachers: Prevalence and outcomes using a threat assessment approach. Teacher and Teacher Education, 87, 1-11. doi.org/10.1016/j.tate.2019.102934

    Internationally, student aggression against teachers is a prevalent problem in schools. Student threat assessment is an emerging violence prevention practice, but its use for threats against teachers has not been investigated. This study examined use of threat assessment for a statewide sample of student threats against teachers (n = 226) compared to threats against other students (n = 1,228). Results indicated that threats against teachers were less prevalent (15.5%) than threats against peers (84.5%). Of threats against teachers, 30% were classified as serious by the school’s threat assessment team and 5.8% were attempted. Implications for school policy and practice and teacher safety are discussed.

    Stohlman, S., Konold, T., & Cornell, D. (2020). Evaluation of threat assessment training for school personnel. Journal of Threat Assessment and Management, 7(1-2), 29-40. https://doi.apa.org/doiLanding?doi=10.1037%2Ftam0000142

    Despite the widespread use of threat assessment in K–12 schools, there is a dearth of research investigating the staff training process. We evaluated the effectiveness of day-long training on the Comprehensive Student Threat Assessment Guidelines (CSTAG) in a sample of 4,666 multidisciplinary school personnel from administration, law enforcement, mental health, teaching, and other groups. Across 100 workshops conducted by 9 trainers, all discipline groups showed large and statistically significant increases in their knowledge of threat assessment from pretest to posttest. On average, participants achieved threat classification accuracy scores of 75% after completing the workshop. Over 95% of participants provided positive evaluations of the workshop and highly endorsed motivation to implement threat assessment in their schools. Overall, these findings support the use of workshop training to prepare multidisciplinary school-based threat assessment teams.

  • Publications 2015-2019

    Burnette, A. G., Konold, T., & Cornell, D. (2019). Grade-level distinctions in student threats of violence. Journal of School Violence, 19:3, 323-335.  https://doi.org/10.1080/15388220.2019.1694031

    Virginia law mandates the use of threat assessment in all public schools, yet there is little research on grade-level differences. This study investigated a statewide sample of 3,282 threats from 1,021 schools. Threats significantly differed across grade level in demographics, characteristics, and outcome. As grade increased, students were more likely to threaten physical assaults (OR = 1.11, p < .001), but less likely to threaten with weapons (OR = 0.95, p < .01). Notably, 1st graders (OR = 2.01, p < .05) were two times more likely to threaten to kill, but 9th graders were more likely to attempt their threats (OR = 1.02, p < .05). These findings highlight the need to consider grade level in evaluating and responding to student threats of violence.

    Crowley, B. Z., Datta, P., Stohlman, S., Cornell, D., & Konold, T. (2019). Authoritative school climate and sexual harassment: A cross-sectional multilevel analysis of student self-reports. School Psychology34, 469.  https://doi.org/10.1037/spq0000303 

    School sexual harassment (SH) is defined as unwelcome behavior of a sexual nature that interferes with a student’s ability to learn. There is an important need for schools to assess the prevalence of SH and its relation to school climate to guide intervention efforts. This study investigated 3 research questions: (a) Is there psychometric support for a 4-item multilevel measure of SH? (b) What is the prevalence of SH in a statewide high school sample, and how does SH vary across gender, grade level, race–ethnicity, and socioeconomic status? (c) Is an authoritative school climate—characterized by strict but fair discipline and supportive teacher–student relationships—associated with lower levels of SH for students? A statewide sample of high school students (N = 62,679) completed a school climate survey that included a new 4-item measure of SH. Results of a multilevel confirmatory factor analysis indicated good fit for a single SH factor at both student and school levels. A multiway analysis of variance demonstrated the high prevalence of SH and variations across demographic groups. Multilevel hierarchical regression analyses indicated that an authoritative school climate accounted for 5.7% of the student-level variance and 38.3% of the school-level variance in SH scores. Routine assessment of SH can help school psychologists bring attention to this underrecognized problem.

    Jia, Y., & Konold, T. (2019). Moving to the next level: Doubly latent multilevel mediation models with a school climate illustration. The Journal of Experimental Education, 1-19. https://doi.org/10.1080/00220973.2019.1675136

    Traditional observed variable multilevel models for evaluating indirect effects are limited by their inability to quantify measurement and sampling error. They are further restricted by being unable to fully separate within- and between-level effects without bias. Doubly latent models reduce these biases by decomposing the observed within-level indicators into within- and between-level latent components. This decomposition simultaneously controls for both measurement error at each level, and sampling error in the aggregation of individual values that serve as indicators of higher-level constructs. We illustrate the usefulness of this approach in investigating the mediating role of aggressive attitudes in associations between reports of school climate and prevalence of teasing and bullying. Teacher perceived justness and support in schools were both found to be indirectly associated with the prevalence of teasing and bullying through aggressive attitudes, and teacher perceived support was found to be directly associated with teasing and bullying.

    Maeng, J., Cornell, D., & Huang, F. (2019). Student threat assessment as an alternative to exclusionary discipline. Journal of School Violence, 19:3, 377-388. https://doi.org/10.1080/15388220.2019.1707682

    Threat assessment has been proposed as a method for schools to respond to student threats of violence that does not rely on exclusionary discipline practices (e.g., suspension, transfer, expulsion, arrest). The present study compared disciplinary consequences for 657 students in 260 schools using the Comprehensive Student Threat Assessment Guidelines (CSTAG) with a comparison group of 661 students in 267 schools using a more general threat assessment approach. The odds that students receiving a threat assessment in CSTAG schools would receive a suspension (OR = 0.59) or law enforcement action (OR = 0.47) were less than those in schools using a general approach. Students in CSTAG schools were expelled at lower rates (0% versus 1.7%) than students in comparison schools. These results indicate that schools using the CSTAG model are less likely to respond to student threats with exclusionary discipline.

    Stohlman, S., & Cornell, D. (2019). An online educational program to increase student understanding of threat assessment. Journal of School Health, 89 (11), 899-906. https://doi.org/10.1111/josh.12827

    Background: Threat assessment is a widely recommended practice used by schools to investigate and respond to student threats of violence; however, students are often reluctant to disclose threats. Methods: We developed an online educational program for students to increase their understanding of threat assessment and the need to report serious threats. We investigated 2 research questions: (1) How are student characteristics of sex, grade level, and ethnicity/race associated with student knowledge of threat assessment and willingness to report threats? (2) Does the program increase knowledge of threat assessment and willingness to report threats? The sample consisted of 2338 students from 6 middle schools and 3 high schools. Results: Prior to program completion, boys were less willing than girls, and older students were less willing than younger students, to report threats. Post‐program questions revealed that the program significantly increased knowledge and willingness to report threats across student groups, with effect sizes (Cohen's d) ranging from small (.30) to large (1.43). Conclusions: This program promotes school safety by teaching students about threat assessment and increasing willingness to report threats. The program is available online for other schools to use.

    Burnette, A. G., Datta, P. & Cornell, D. G. (2018). The distinction between transient and substantive student threats. Journal of Threat Assessment and Management, 5, 4-20. http://psycnet.apa.org/record/2017-56103-001

    Many schools across North America have adopted student threat assessment as a violence prevention strategy. The Virginia Student Threat Assessment Guidelines (VSTAG) is a threat assessment model that emphasizes distinguishing between substantive threats that are serious and transient threats that are not serious. This retrospective study investigated the interrater reliability and criterion-related validity of this distinction in a sample of 844 student threat cases from 339 Virginia public schools. To assess interreliability for the transient versus substantive distinction, research coders independently classified a subsample of 148 narratives, achieving classification agreement with schools of 70% (κ = .53). Logistic regression analyses examined transient and substantive threat differences in threat characteristics and outcomes. Threats were more likely to be classified as substantive when they included warning behaviors (e.g., history of violence, weapon use, leakage, etc.), were made by older students, mentioned the use of a bomb or a knife, and involved threats to harm self as well as others. Although only 2.5% of threats were attempted, substantive threats were 36 times more likely to be attempted than transient threats. Substantive threats were more likely to result in out-of-school suspension, change in school placement, and/or legal action. Overall, these findings supported the transient/substantive distinction, but indicated some training needs for school teams.

    Burnette, A.G., Huang, F., Maeng, J.L., & Cornell, D. (2018). School threat assessment versus suicide assessment: Statewide prevalence and case characteristics. Psychology in the Schools, 56(3), 378-392. https://doi.org/10.1002/pits.22194

    Threat assessment is a violence prevention strategy used to investigate and respond to threats to harm others. In 2013, Virginia mandated the use of threat assessment teams for threats to self and to others, effectively subsuming suicide assessment with threat assessment and raising questions about the distinction between the two practices. In a statewide sample of 2,861 cases from 926 schools, there were more threats to self (60%) than others (35%), with only 5% involving threats to both self and others. Threats to self were more likely to be made by females (odds ratio [OR] = 3.38) and students with fewer prior disciplinary actions (OR = 0.48). Threats to self were much less likely to involve a weapon (OR = 0.07), but more likely to be attempted (OR = 1.50) and result in mental health services (OR = 2.96). They were much less likely to result in out‐of‐school suspensions (OR = 0.07), legal action (OR = 0.17), and/or changes in placement (OR = 0.53). Overall, these findings support a clear distinction between suicide and threat assessment.

    Cornell, D., & Maeng, J. (2018). Statewide implementation of threat assessment in Virginia K-12 schools. Contemporary School Psychology, 22, 116-124. https://doi.org/10.1007/s40688-017-0146-x

    In 2013, VA became the first state to mandate the use of threat assessment teams in its K-12 public schools. We provide an account of the development and adaptation of threat assessment as a school safety practice and research on the Virginia Student Threat Assessment Guidelines in VA schools. We describe the state law and the question of whether suicide assessment should be considered a form of threat assessment. We then describe research on the statewide implementation of threat assessment and summarize results indicating overall positive outcomes for schools who are actively engaged in threat assessment, but qualitative findings from a needs assessment identified team training gaps as well as a need to orient the larger school community to the threat assessment process. We describe a series of online programs to educate students, parents, teachers, and other school staff about the threat assessment process. In conclusion, this paper presents some lessons learned in the statewide implementation of threat assessment as a safe and effective violence prevention strategy.

    Cornell, D., Maeng, J., Burnette, A.G., Jia, Y., Huang, F., Konold, T., Datta, P., Malone, M., Meyer, P. (2018). Student threat assessment as a standard school safety practice: Results from a statewide implementation study. School Psychology Quarterly, 33, 213-222. http://dx.doi.org/10.1037/spq0000220

    Threat assessment has been widely endorsed as a school safety practice, but there is little research on its implementation. In 2013, Virginia became the first state to mandate student threat assessment in its public schools. The purpose of this study was to examine the statewide implementation of threat assessment and to identify how threat assessment teams distinguish serious from nonserious threats. The sample consisted of 1,865 threat assessment cases reported by 785 elementary, middle, and high schools. Students ranged from pre-K to Grade 12, including 74.4% male, 34.6% receiving special education services, 51.2% White, 30.2% Black, 6.8% Hispanic, and 2.7% Asian. Survey data were collected from school-based teams to measure student demographics, threat characteristics, and assessment results. Logistic regression indicated that threat assessment teams were more likely to identify a threat as serious if it was made by a student above the elementary grades (odds ratio 0.57; 95% lower and upper bound 0.42–0.78), a student receiving special education services (1.27; 1.00–1.60), involved battery (1.61; 1.20–2.15), homicide (1.40; 1.07–1.82), or weapon possession (4.41; 2.80–6.96), or targeted an administrator (3.55; 1.73–7.30). Student race and gender were not significantly associated with a serious threat determination. The odds ratio that a student would attempt to carry out a threat classified as serious was 12.48 (5.15–30.22). These results provide new information on the nature and prevalence of threats in schools using threat assessment that can guide further work to develop this emerging school safety practice.

    Cornell, D., Maeng, J., Huang, F., Shukla, K., & Konold, T. (2018). Racial/ethnic parity in disciplinary consequences using student threat assessment. School Psychology Review, 47, 183-195. https://doi.org/10.17105/SPR-2017-0030.V47-2

    School psychologists are frequently called upon to assess students who have made verbal or behavioral threats of violence against others, a practice commonly known as threat assessment. One critical issue is whether the outcomes of a threat assessment generate the kind of racial disparities widely observed in school disciplinary practices. In 2013, Virginia became the first state to mandate threat assessment teams in all public schools. This study examined the disciplinary consequences for 1,836 students who received a threat assessment in 779 Virginia elementary, middle, and high schools during the 2014–2015 school year. Multilevel logistic regression models found no disparities among Black, Hispanic, and White students in out-of-school suspensions, school transfers, or legal actions. The most consistent predictors of disciplinary consequences were the student's possession of a weapon and the team classification of the threat as serious. We discuss possible explanations for the absence of racial/ethnic disparities in threat assessment outcomes and cautiously suggest that the threat assessment process may reflect a generalizable pathway for achieving parity in school discipline.

    Huang, F. & Cornell D. (2017). Student attitudes and behaviors as explanations for the Black-White suspension gap. Children and Youth Services Review, 73, 298-308.

    Purpose: Although studies have documented that Black students receive out-of-school suspensions (OSS) at much higher rates than White students, few studies have investigated possible explanations for this disparity. The differential involvement hypothesis suggests that disproportionate sanctioning may be a function of racial differences in student misbehavior or characteristics that predispose them to misbehavior. Method: Suspension data, risk behaviors, and aggressive attitudes from self-report surveys were collected from a statewide sample of 38,398 students attending 236 racially-diverse high schools. A series of school fixed-effect logistic and linear regression models were used to test behavioral and attitudinal forms of the differential involvement hypothesis. Results: Racial differences in self-reported suspension could not be explained by different behavioral reasons for suspension (such as fighting, threatening others, and substance possession), by involvement in high risk behaviors of fighting, bullying, carrying a weapon, consuming alcohol, or using marijuana, or by aggressive attitudes that lead to hostile behavior. Conclusions: Overall, these findings do not support the differential involvement hypothesis and although they do not establish the presence of bias, they strengthen concern that racial disparities are likely the result of differential decisions by school authorities.

    Leuschner, V.,  Fiedler, N., Schultze, M., Ahlig, N., Göbel, K. Sommer, F., Scholl, J. Cornell, D., & Scheithauer, H. (2017). Prevention of targeted school violence by responding to students’ psychosocial crises: The NETWASS Program. Child Development, 88, 68-82.

    The standardized, indicated school-based prevention program “Networks Against School Shootings (NETWASS) combines a threat assessment approach with a general model of prevention of emergency situations in schools through early intervention in student psychosocial crises and training teachers to recognize warning signs of targeted school violence. An evaluation study in 98 German schools with 3,473 school staff participants used a quasi-experimental comparison group design with three measurement points (pre, post, 7-months-follow-up) with school randomly allocated to implementation conditions. The study found increases in teachers’ expertise and evaluation skills, enhanced abilities to identify students experiencing a psychosocial crisis, and positive secondary effects (e.g. teacher-student interaction, feelings of safety).

    Berg, J., & Cornell, D. (2016). Authoritative school climate, aggression toward teachers, and teacher distress in middle school. School Psychology Quarterly, 31, 122-139. http://dx.doi.org/10.1037/spq0000132

    Aggression toward teachers is linked to burnout and disengagement from teaching, but a positive school climate may reduce aggression and associated teacher distress. Using authoritative school climate theory, the study examined whether schools with high disciplinary structure and student support were associated with less aggression and less distress. The sample of 9,134 teachers in 389 middle schools came from the Virginia Secondary School Climate Survey, a statewide survey administered to all public schools with 7th and 8th grade enrollment. The majority of teachers (75%) were female. More than half (53%) reported that they had more than 10 years of teaching experience; 23% reported 6 to 10 years; 24% reported 1 to 5 years. Students reported on the degree to which their schools were structured and supportive. Teachers reported on their experiences of aggression by students, their level of distress, and their feelings of safety. Staff-related infractions computed from Department of Education records were also used. Multilevel modeling revealed that teachers in authoritative schools experienced less aggression and felt safer and less distressed. Lower aggression by students mediated the association between more authoritative schools and lower dis- tress such that more structured and supportive schools had greater teacher safety and, in turn, less distress. The findings support the idea that more structured and supportive schools relate to greater safety for teachers and, in turn, less distress. Research limitations and implications for practice are discussed. 

    Cornell, D., & Huang, F. (2016). Authoritative school climate and high school student risk behavior. A cross-sectional multi-level analysis of student self-reports. Journal of Youth and Adolescence, 45, 2246-2259, doi: 10.1007/s10964-016-0424-3

    Many adolescents engage in risk behaviors such as substance use and aggression that jeopardize their healthy development. This study tested the hypothesis that an authoritative school climate characterized by strict but fair discipline and supportive teacher-student relationships is conducive to lower risk behavior for high school students. Multilevel logistic regression models were used to analyze cross-sectional, student-report survey data from a statewide sample of 47,888 students (50.6% female) in 319 high schools. The students included ninth (26.6%), tenth (25.5%), eleventh (24.1%) and twelfth (23.8%) grade with a racial/ethnic breakdown of 52.2% White, 18.0% Black, 13.1% Hispanic, 5.9% Asian, and 10.8% reporting another or two or more race/ethnicities. Schools with an authoritative school climate had lower levels of student-reported alcohol and marijuana use; bullying, fighting, and weapon carrying at school; interest in gang membership; and suicidal thoughts and behavior. These results controlled for demographic variables of student gender, race, grade, and parent education level as well as school size, percentage of minority students, and percentage of low income students. Overall, these findings add new evidence that an authoritative school climate is associated with positive student outcomes.

    Cornell, D., Shukla, K., & Konold, T. (2016). Authoritative school climate and student academic engagement, grades, and aspirations in middle and high schools. AERA Open, 2, 1-18, doi: 10.1177/2332858416633184

    This study tested the theory that an authoritative school climate characterized by disciplinary structure and student support is conducive to positive academic outcomes for middle and high school students. Multilevel multivariate modeling at student and school levels was conducted using school surveys completed by statewide samples of 39,364 students in Grades 7 and 8 in 423 middle schools and 48,027 students in Grades 9 through 12 in 323 high schools. Consistent with authoritative school climate theory, both higher disciplinary structure and student support were associated with higher student engagement in school, higher course grades, and higher educational aspirations at the student level in both samples. At the school level, higher disciplinary structure was associated with higher engagement, and higher student support was associated with higher engagement and grades in both samples. Overall, these findings add new evidence that an authoritative school climate is conducive to student academic success in middle and high schools. 

    Datta, P., Cornell, D., & Huang, F. (2016). Aggressive attitudes and prevalence of bullying bystander behaviors in middle schools. Psychology in the Schools, 53, 804-816. doi: 10.1002/pits.21944

    Separate lines of research find that pro-aggressive attitudes promote peer aggression and that bystanders play a pivotal role in deterring or facilitating bullying behavior. The current study hypothesized that pro-aggressive attitudes in middle school would deter students from standing up to bullying and encourage them to reinforce bullying behavior. Middle school students (n = 28,765) in 423 schools completed a statewide school climate survey that included an aggressive attitudes scale and their bystander response to a recent episode of bullying, which was categorized as upstanding, reinforcing, or passive. Multilevel logistic regressions indicated that higher aggressive attitudes were associated with less upstanding behavior at the school level and less upstanding behavior and more reinforcing behavior at the individual level, while controlling for other school and student demographic variables. These findings suggest that antibullying programs might address student attitudes toward aggression as a means of boosting positive bystander intervention. 

    Huang, F., Eklund, K., & Cornell, D. (2016). Authoritative school climate, number of parents at home, and academic achievement. School Psychology Quarterly. Advance online publication. http://dx.doi.org/10.1037/spq0000182

    School climate is widely recognized as an important factor in promoting student academic achievement. The current study investigated the hypothesis that a demanding and supportive school climate, based on authoritative school climate theory, would serve as a protective factor for students living with one or no parents at home. Using a statewide sample of 56,508 middle school students from 415 public schools in one state, results indicated that student perceptions of disciplinary structure, academic demandingness, and student support all had positive associations with student self-reported grade point average (GPA). In addition, findings showed that academic expectations and student support were more highly associated with GPA for students not living with any parent. Implications for policy and practice are discussed. 

    Jia Y., Konold R. T., Cornell D., & Huang F. (2016) The impact of validity screening on associations between self-reports of bullying victimization and student outcomes. Educational and Psychological Measurement, 0, 1-23, doi: 10.1177/0013164416671767

    Self-report surveys are widely used to measure adolescent risk behavior and academic adjustment, with results having an impact on national policy, assessment of school quality, and evaluation of school interventions. However, data obtained from self-reports can be distorted when adolescents intentionally provide inaccurate or careless responses. The current study illustrates the problem of invalid respondents in a sample (N = 52,012) from 323 high schools that responded to a statewide assessment of school climate. Two approaches for identifying invalid respondents were applied, and contrasts between the valid and invalid responses revealed differences in means, prevalence rates of student adjustment, and associations among reports of bullying victimization and student adjustment outcomes. The results lend additional support for the need to screen for invalid responders in adolescent samples. 

    Konold, T.R. (2016). A multilevel MTMM approach to estimating the influences of contextual factors on trait and informant based method effects in assessments of school climate. Journal of Psychoeducational Assessment. doi: 0734282916683286

    School level contextual factors have been found to influence reports of school climate. The purpose of the current study was to evaluate the extent to which these associations are related to the school climate traits being measured or the methods (i.e., informants) used to obtain them. Data from a multilevel MTMM design in which structurally different and interchangeable students (N = 45,641) and teachers (N = 12,808), residing within 302 high schools, responded to items measuring four dimensions of school climate were evaluated through a multilevel CT - CM latent analysis that allowed for the estimation of both school level trait and informant based method factors. The resulting trait and method factors were regressed on several school level contextual variables. Results indicated that the percent of students receiving FRPM in schools was associated with both school climate traits and informant based method factors, school size and the percentage of minority students in schools were associated with some traits, and school size was associated with student method effects. Findings support the use of controlling for school level contextual factors in school climate research. 

    Konold, T., Cornell, D., Shukla, K., & Huang, F. (2016). Racial/ethnic differences in perceptions of school climate and its association with student engagement and peer aggression. Journal of Youth and Adolescence. Advance online publication. doi: 10.1007/s10964-016-0576-1

    Research indicates that a positive school climate is associated with higher levels of student engagement and lower rates of peer aggression. However, less attention has been given to whether such findings are consistent across racial/ethnic groups. The current study examined whether Black, Hispanic, and White high school students differed in their perceptions of school climate, student engagement, and peer aggression as measured by the Authoritative School Climate survey. In addition, the study tested whether the associations between school climate and both student engagement and peer aggression varied as a function of racial/ethnic group. The sample consisted of 48,027 students in grades 9–12 (51.4 % female; 17.9 % Black, 10.5 % Hispanic, 56.7 % White, and 14.9 % other) attending 323 high schools. Regression models that contrasted racial/ethnic groups controlled for the nesting of students within schools and used student covariates of parent education, student gender, and percentage of schoolmates sharing the same race/ethnicity, as well as school covariates of school size and school percentage of students eligible for free- or reduced-price meals. Perceptions of school climate differed between Black and White groups, but not between Hispanic and White groups. However, race/ethnicity did not moderate the associations between school climate and either engagement or peer aggression. Although correlational and cross-sectional in nature, these results are consistent with the conclusion that a positive school climate holds similar benefits of promoting student engagement and reducing victimization experiences across Black, Hispanic, and White groups.

    Konold, T.R., & Shukla, K. (2016). Estimating school climate traits across multiple informants: An illustration of a multi-trait multi-method validation through latent variable modeling. Educational Assessment. Online advanced publication. http://dx.doi.org/10.1080/10627197.2016.1271705

    The use of multiple informants is common in assessments that rely on the judgments of others. However, ratings obtained from different informants often vary as a function of their perspectives and roles in relation to the target of measurement, and causes unrelated to the trait being measured. We illustrate the usefulness of a latent variable multilevel MTMM measurement model for extracting trait factors from reports of school climate obtained by students (N = 45,641) and teachers (N = 12,808) residing within 302 high schools. We then extend this framework to include assessments of linkages between the resulting trait factors and potential outcomes that might be used for addressing questions of substantive interest or providing evidence of concurrent validity. The approach is illustrated with data obtained from student and teacher reports of two dimensions of school climate, student engagement, and the prevalence of teasing and bullying in their schools. 

    Malone, M., Cornell, D., & Shukla, K. (2016). Association of grade configuration with school climate for 7th and 8th grade students. School Psychology Quarterly. Advance online publication. doi: 10.1037/spq0000174

    ducational authorities have questioned whether middle schools provide the best school climate for 7th and 8th grade students, and proposed that other grade configurations such as K–8th grade schools may provide a better learning environment. The purpose of this study was to compare 7th and 8th grade students’ perceptions of 4 key features of school climate (disciplinary structure, student support, student engagement, and prevalence of teasing and bullying) in middle schools versus elementary or high schools. Multilevel multivariate modeling in a statewide sample of 39,036 7th and 8th grade students attending 418 schools revealed that students attending middle schools had a more negative perception of school climate than students in schools with other grade configurations. Seventh grade students placed in middle schools reported lower disciplinary structure and a higher prevalence of teasing and bullying in comparison to those in elementary schools. Eighth grade students in middle schools reported poorer disciplinary structure, lower student engagement, and a higher prevalence of teasing and bullying compared to those in high schools. These findings can guide school psychologists in identifying aspects of school climate that may be troublesome for 7th and 8th grade students in schools with different grade configurations.

    Shukla, K., Konold, T., & Cornell, D. (2016). Profiles of student perceptions of school climate: Relations with risk behaviors and academic outcomes. American Journal of Community Psychology, 57, 291-307. doi: 10.1002/ajcp.12044

    School climate has been linked to a variety of positive student outcomes, but there may be important within- school differences among students in their experiences of school climate. This study examined within-school heterogeneity among 47,631 high school student ratings of their school climate through multilevel latent class modeling. Student profiles across 323 schools were generated on the basis of multiple indicators of school climate: disciplinary structure, academic expectations, student willingness to seek help, respect for students, affective and cognitive engagement, prevalence of teasing and bullying, general victimization, bullying victimization, and bullying perpetration. Analyses identified four meaningfully different student profile types that were labeled positive climate, medium climate-low bullying, medium climate-high bullying, and negative climate. Contrasts among these profile types on external criteria revealed meaningful differences for race, grade-level, parent education level, educational aspirations, and frequency of risk behaviors.

    Cornell, D. (2015). Our schools are safe: Challenging the misperception that schools are dangerous places. American Journal of Orthopsychiatry, 85, 217-220. http://dx.doi.org/10.1037/ort0000064

    Massive public attention to school shootings has created the misperception that schools are dangerous places, even though crime statistics show that schools are one of the safest places in the United States. The fear of school shootings has caused many school systems to divert their budgets to excessive building security measures and adopt dubious crisis response plans. School disciplinary practices have shifted toward the criminalization of student misbehavior and a zero tolerance philosophy that fails to improve school safety and results in high rates of student suspensions and dropouts. The use of a threat assessment approach to evaluate individual student behavior in context and resolve conflicts and problems before they escalate into violence is one promising alternative that has been adopted statewide in Virginia public schools. School safety should focus on the everyday problems of bullying and fighting, and apply public health principles of primary and secondary prevention using well-established psychological interventions.

    Cornell, D., & Limber, S. (2015). Law and policy on the concept of bullying at school. American Psychologist, 70, 333-343.

    The nationwide effort to reduce bullying in U.S. schools can be regarded as part of larger civil and human rights movements that have provided children with many of the rights afforded to adult citizens, including protection from harm in the workplace. Many bullied children find that their schools are hostile environments, but civil rights protections against harassment apply only to children who fall into protected classes, such as racial and ethnic minorities, students with disabilities, and victims of gender harassment or religious discrimination. This article identifies the conceptual challenges that bullying poses for legal and policy efforts, reviews judicial and legislative efforts to reduce bullying, and makes some recommendations for school policy. Recognition that all children have a right to public education would be one avenue for broadening protection against bullying to all children.

    Cornell, D., Shukla, K., & Konold, T. (2015). Peer victimization and authoritative school climate: A multilevel multivariate approach. Journal of Educational Psychology, 107, 1186-1201. http://dx.doi.org/10.1037/edu0000038

    School climate is widely recognized as an important influence on peer victimization in schools. The purpose of this study is to examine how authoritative school climate theory provides a framework for conceptualizing 2 key features of school climate—disciplinary structure and student support—that are associated with 3 measures of peer victimization. Multilevel multivariate modeling in a statewide sample of 39,364 7th- and 8th-grade students attending 423 schools revealed meaningful associations at both the student and school levels of analysis. Higher disciplinary structure was associated with lower levels of prevalence of teasing and bullying, bullying victimization, and general victimization. Higher student support was associated with lower prevalence of teasing and bullying and general victimization. Overall, these findings add new evidence to the theory that an authoritative school climate is conducive to lower peer victimization.

    Heilbrun, A., Cornell, D., & Lovegrove, P. (2015). Principal attitudes and racial disparities in school suspensions. Psychology in the Schools, 52, 489-499.

    Zero tolerance school discipline practices have been associated with a national increase in suspensions, a practice that has had a disproportionate negative impact on Black students. The present study investigated an association between principal attitudes toward zero tolerance and suspension rates for White and Black students in 306 Virginia high schools. Black suspension rates were more than double White suspension rates. Regression analyses controlling for student poverty and school enrollment showed that principal endorsement of zero tolerance was moderately associated with suspension rates for both White and Black students, but was not associated with the size of the racial disparity. Paired-samples t tests showed statistically significant differences in the types of offenses that resulted in suspensions, with Black students significantly more likely to be suspended for disruptive offenses and White students more likely to be suspended for alcohol- and drug-related offenses.

    Huang, F. & Cornell, D. (2015). The impact of definition and question order on the prevalence of bullying victimization using student self-reports. Psychological Assessment, 27, 1484-1493. http://dx.doi.org/10.1037/pas0000149

    Accurate measurement is essential to determining the prevalence of bullying and evaluating the effectiveness of intervention efforts. The most common measurement approach is through anonymous self-report surveys, but previous studies have suggested that students do not adhere to standard definitions of bullying and may be influenced by the order of questions about types of victimization. In the current study, we have presented findings from 2 randomized experiments designed to determine (a) the impact of using or not using a definition of bullying and (b) asking about general versus specific types of bullying victimization and how the order of these questions affects victimization-prevalence rates. The study was conducted using a sample of 17,301 students attending 119 high schools. Findings indicate that the use of a definition had no impact on prevalence rates, but asking specific bullying-victimization questions (e.g., “I have been verbally bullied at school”) prior to general bullying-victimization questions (e.g., “I have been bullied at school”), resulted in a 29–76% increase in victimization-prevalence rates. Results suggest that surveys that ask general-to-specific bullying-victimization questions, such as those found in national and international surveys, may be underreporting bullying victimization. 

    Huang, F., & Cornell, D. (2015). Multilevel factor structure, concurrent validity, and test-retest reliability of the high school teacher version of the Authoritative School Climate Survey. Journal of Psychoeducational Assessment, 34, 536-549. doi: 10.1177/0734282915621439

    Although school climate has long been recognized as an important factor in the school improvement process, there are few psychometrically supported measures based on teacher perspectives. The current study replicated and extended the factor structure, concurrent validity, and test–retest reliability of the teacher version of the Authoritative School Climate Survey (ASCS) using a statewide sample of high school teachers. Multilevel confirmatory factor analyses based on surveys completed by 12,808 high school teachers from 302 schools found that factors of disciplinary structure and student support were associated to varying degrees with the teacher reports of the prevalence of student teasing and bullying and student engagement. These findings provide some empirical support for the use of the teacher version of the ASCS in high schools.

    Huang, F., & Cornell, D. (2015). Question order affects the measurement of bullying victimization. Educational and Psychological Measurement, 76, 724-740, doi: 10.1177/0013164415622664

    Bullying among youth is recognized as a serious student problem, especially in middle school. The most common approach to measuring bullying is through student self- report surveys that ask questions about different types of bullying victimization. Although prior studies have shown that question-order effects may influence participant responses, no study has examined these effects with middle school students. A randomized experiment (n = 5,951 middle school students) testing the question- order effect found that changing the sequence of questions can result in 45% higher prevalence rates. These findings raise questions about the accuracy of several widely used bullying surveys. 

    Huang, F., & Cornell, D. (2015). Using multilevel factor analysis with clustered data: Investigating the factor structure of the Positive Values Scale. Journal of Psychoeducational Assessment, 34, 3-14. doi: 10.1177/0734282915570278

    Advances in multilevel modeling techniques now make it possible to investigate the psychometric properties of instruments using clustered data. Factor models that overlook the clustering effect can lead to underestimated standard errors, incorrect parameter estimates, and model fit indices. In addition, factor structures may differ depending on the level of analysis. The current study illustrates the application of multilevel factor analytic techniques using a large statewide sample of middle school students (n = 39,364) from 423 schools. Both multilevel exploratory and confirmatory factor analyses were used to investigate the factor structure of the Positive Values Scale (PVS) as part of a school climate survey. Results showed that for the PVS, a two-correlated factor model at Level 1 and a one-factor model at Level 2 best fit the data. Implications and guidance for applied researchers are discussed. 

    Huang, F., Cornell, D., Konold, T., Meyer, P., Lacey, A., Nekvasil, E., Heilbrun, A., & Shukla, K. (2015). Multilevel factor structure and concurrent validity of the teacher version of the Authoritative School Climate Survey. Journal of School Health, 85, 843-851.

    Background: School climate is well-recognized as an important influence on student behavior and adjustment to school, but there is a need for theory-guided measures that make use of teacher perspectives. Authoritative school climate theory hypothesizes that a positive school climate is characterized by high levels of disciplinary structure and student support. Methods: A teacher version of the Authoritative School Climate Survey was administered to a statewide sample of 9,099 7th and 8th grade teachers from 366 schools. The study used exploratory and multilevel confirmatory factor analyses (MCFA) that accounted for the nested data structure and allowed for the modeling of the factor structures at two levels. Results: MCFA conducted on both an exploratory (n = 4,422) and a confirmatory sample (n = 4,677) showed good support for all of the factor structures investigated. An overall model that considered all factor correlations at two levels simultaneously found that schools with greater levels of disciplinary structure and student support had higher student engagement, less teasing and bullying, and lower student aggression toward teachers. Conclusions: The teacher version of the Authoritative School Climate Survey can be used to assess two key domains of school climate and associated measures of student engagement and aggression toward peers and teachers.

    Jia, Y., Konold, T., & Cornell, D. (2015). Authoritative school climate and high school dropout rates. School Psychology Quarterly, 31, 289-303. http://dx.doi.org/10.1037/spq0000139

    This study tested the association between school-wide measures of an authoritative school climate and high school dropout rates in a statewide sample of 315 high schools. Regression models at the school level of analysis used teacher and student measures of disciplinary structure, student support, and academic expectations to predict overall high school dropout rates. Analyses controlled for school demographics of school enrollment size, percentage of low-income students, percentage of minority students, and urbanicity. Consistent with authoritative school climate theory, moderation anal- yses found that when students perceive their teachers as supportive, high academic expectations are associated with lower dropout rates.

    Konold, T., & Cornell, D. (2015). Measurement and structural relations of an Authoritative School Climate model: A multi-level latent variable investigation. Journal of School Psychology, 53, 447-461. http:/dx.doi.org/10.1016/j.jsp.2015.09.001

    This study tested a conceptual model of school climate in which two key elements of an authoritative school, structure and support variables, are associated with student engagement in school and lower levels of peer aggression. Multilevel multivariate structural modeling was conducted in a statewide sample of 48,027 students in 323 public high schools who completed the Authoritative School Climate Survey. As hypothesized, two measures of structure (Disciplinary Structure and Academic Expectations) and two measures of support (Respect for Students and Willingness to Seek Help) were associated with higher student engagement (Affective Engagement and Cognitive Engagement) and lower peer aggression (Prevalence of Teasing and Bullying) on both student and school levels of analysis, controlling for the effects of school demographics (school size, percentage of minority students, and percentage of low income students). These results support the extension of authoritative school climate model to high school and guide further research on the conditions for a positive school climate.

    Konold, T., & Cornell, D. (2015). Multilevel, multitrait - multimethod latent analysis of structurally different and interchangeable raters of school climate. Psychological Assessment, 27, 1097-1109. doi: 10.1037/pas0000098

    Informant-based systems of assessment are common platforms for measuring a variety of educational and psychological constructs where the use of multiple informants is considered best practice. In many instances, structurally different informant types (e.g., students and teachers) are solicited on the basis of their unique roles with the target of measurement. The use of multiple informants provides an opportunity to evaluate the degree to which the obtained ratings are influenced by the trait of focus and extraneous sources that can be attributed to the rater. Data from a multilevel multitrait–multimethod design in which students (N 35,565) and teachers (N 9,112), from 340 middle schools, responded to items measuring 3 dimensions of school climate were evaluated through a multilevel correlated trait– correlated method latent variable model. Results indicated that ratings of school climate obtained by students and teachers demonstrated high levels of convergent validity, and that school-level ratings obtained by students and teachers were equitable in the assessment of teasing and bullying. Student ratings of support and structure yielded somewhat stronger evidence of convergent validity than ratings obtained by teachers as revealed by their respective trait factor loadings. This was explained in part by the higher levels of common method effects that were observed for teachers.

    Lacey, A., Cornell, D., & Konold, T. (2015). The relations between teasing and bullying and middle school standardized exam performance. The Journal of Early Adolescence, 37,192-221. doi: 0272431615596428.

    This study examined the relations between the schoolwide prevalence of teasing and bullying and schoolwide academic performance in a sample of 271 Virginia middle schools. In addition, the study examined the mediating effects of student engagement. A three-step sequence of path models investigated associations between schoolwide prevalence of teasing and bullying and state-mandated Standards of Learning test pass rates, with effects examined both directly and indirectly through student engagement while controlling for important school-level characteristics. Separate models were examined for two 7th grade and four 8th grade tests. Results indicated that higher levels of both teacher and student perceptions of schoolwide teasing and bullying were significantly associated with lower achievement pass rates and student engagement. The relationship between perceptions of schoolwide teasing and bullying and achievement was partially mediated by student engagement. These findings bring new support for the need for schoolwide interventions to reduce teasing and bullying among middle schools students.

    Millspaugh, S. B., Cornell, D. G., Huang, F. L., & Datta, P. (2015). Prevalence of aggressive attitudes and willingness to report threats in middle school. Journal of Threat Assessment and Management, 2, 11-22. doi: http://dx.doi.org/10.1037/tam0000031

    Violence prevention strategies such as threat assessment rely on information from students; however, students are often unwilling to report threats of violence to school authorities. The current study investigated the hypothesis that middle school students are less likely to report threats of violence when they perceive aggressive behavior as a source of status and popularity among their peers. Our statewide sample consisted of 39,364 7th and 8th graders who completed school climate surveys in 423 schools. Students completed a measure of aggressive attitudes and were asked how much they agreed or disagreed with 2 statements concerning threats of violence: (a) “If another student brought a gun to school, I would tell one of the teachers or staff at school,” and (b) “If another student talked about killing someone, I would tell one of the teachers or staff at school.” Multilevel logistic regression analyses, which controlled for student and school demographics, found that higher levels of aggressive attitudes at both the school and student level were associated with a lower likelihood of reporting threat behavior. 

    Nekvasil, E., & Cornell, D. (2015). Student threat assessment associated with positive school climate in middle schools. Journal of Threat Assessment and Management 2, 98-113. http://dx.doi.org/10.1037/tam0000038

    Authorities in law enforcement and education have recommended the use of threat assessment to prevent violence, but few studies have examined its usefulness in middle schools. This retrospective, quasi-experimental study compared middle schools that use the Virginia Student Threat Assessment Guidelines (Cornell & Sheras, 2006; N = 166) to schools that either do not use threat assessment (N = 119) or use an alternative model of threat assessment (school- or district-developed; N = 47). Based on school records, schools using the Virginia Guidelines reported lower short-term suspension rates than both groups of schools. According to a statewide school climate survey, schools using the guidelines also had fairer discipline and lower levels of student aggressive behaviors, as reported by students. Finally, teachers reported feeling safer in schools using the Virginia Guidelines, as opposed to both groups of schools. Additional analyses of school records found that the number of years a school used the Virginia Guidelines was associated with lower long-term suspension rates, student reports of fairer discipline, and lower levels of student aggressive behaviors. All analyses controlled for school size, minority composition, and socioeconomic status of the student body. These findings suggest that use of a threat assessment approach to violence prevention is associated with lower levels of student aggression and a more positive school climate.

    Nekvasil, E., Cornell, D., & Huang, F. (2015). Prevalence and offense characteristics of multiple casualty homicides: Are schools at higher risk than other locations? Psychology of Violence, 5, 236-245.

    Objective: In light of public concern about school shootings, this study examined the prevalence and offense characteristics of multiple casualty homicides across locations. Method: We used the FBI’s National Incident Based Reporting System (NIBRS) to examine 18,873 homicide incidents involving 25,180 victims who were either killed or injured from 2005 through 2010. Results: Multiple casualty homicides were surprisingly common events, with approximately 22% of homicide incidents involving two or more victims. Multiple casualty homicides were much more common in residences (47%) versus schools (0.8%), but homicides in residences tended to have one victim (78%) rather than multiple victims (22%), whereas homicides in schools were about equally likely to have one victim (57%) or multiple victims (43%). Multiple homicides were more likely to involve firearms than weapons such as knives or blunt objects. Finally, there were statistical differences in offense characteristics for homicides with one, two, and three victims. Conclusion: These findings suggest that the public perception that schools are a high-risk location for homicides is inaccurate. Although concern about school shootings is understandable, the larger problem of multiple casualty shootings is more common in other locations which do not receive comparable media attention.

  • Publications 2010-2014

    Baly, M., Cornell, D., & Lovegrove, P., (2014). A longitudinal comparison of peer- and self-reports of bullying victimization across middle school. Psychology in the Schools, 51, 217-247. doi: 10.1002/pits.21747

    Cross-sectional studies indicate how many students are victims of bullying at a single time, but do not tell us whether the same students continue to be bullied or whether there is a cumulative impact of bullying over time. This study examined the longitudinal stability and the cumulative impact of victimization in a sample of 382 students assessed in the fall and the spring of grades 6, 7, and 8.Victimization assessed by both self- and peer-reports indicated substantial variability in who was bullied, with nearly 51% of students reporting bullying victimization during at least one of the six assessments. The cumulative impact of victimization over three years was demonstrated on grade 8 outcome measures of absences, disciplinary infractions, suspensions, grade point averages (GPA), standardized test scores, reports of youth risk behavior, and perceptions of school climate. This study provides new information about the cumulative impact of peer- and self-reported bullying across middle school.

    Cornell, D., G., Lovegrove, P. J., & Baly, M. W. (2014). Invalid survey response patterns among middle school students. Psychological Assessment, 26, 277-287. doi: 10.1037/a0034808

    Student surveys are widely used to assess student risk behavior, bullying, and school climate in middle schools; however, because such surveys are usually conducted on an anonymous basis, little is known about the validity of student reports using external, independent criteria. This longitudinal study examined the response patterns of 382 middle school students who completed confidential (not anonymous) self-report surveys each fall and spring for three years of middle school (grades 6-8). Approximately 10% of students in each wave indicated on validity screening questions that they were either not telling the truth or paying attention (termed “invalid responders”). A repeated measures latent class analysis found that students could be classified into a large group (64%) that were never flagged by the validity questions and a smaller group (36%) that occasionally reported not telling the truth or not paying attention. Hierarchical linear modeling analyses found that invalid responding to validity questions was associated with higher self-reported rates of risk behavior and more negative perceptions of school climate. Based on independent criteria from school records, invalid responding students were more likely to be referred for disciplinary infractions than other students. This study provides new information about student survey validity and appears to be the first to identify characteristics of students who generate invalid response patterns.

    Huang, F., Cornell, D., & Konold, T. (2014). Aggressive attitudes in middle schools: A factor structure and criterion-related validity study. Assessment. Advance online publication: doi: 1073191114551016

    Student attitudes toward aggression have been linked to individual aggressive behavior, but the relationship between school-wide normative beliefs about aggression and aggressive behavior poses some important measurement challenges that have not been adequately examined. The current study investigated the factor structure, measurement invariance, and criterion-related validity of a six-item Aggressive Attitudes scale using a large sample of seventh- and eighth-grade students (n = 39,364) from 423 schools. Analytic procedures accounted for the frequently ignored modeling problems of clustered and ordinal data to provide more reliable and accurate model estimates and standard errors. The resulting second-order factor structure of the Aggressive Attitudes scale demonstrated measurement invariance across gender, grade, and race/ethnicity groups. Criterion-related validity was supported with eight student- and school-level indices of aggressive behavior.

    Huang, F., Cornell, D., Konold, T., Meyer, P., Lacey, A., Nekvasil, E., Heilbrun, A., & Shukla, K. (2014). Multilevel factor structure and concurrent validity of the teacher version of the Authoritative School Climate Survey. Journal of School Health, 85, 843-859.

    School climate is well recognized as an important influence on student behavior and adjustment to school, but there is a need for theory-guided measures that make use of teacher perspectives. Authoritative school climate theory hypothesizes that a positive school climate is characterized by high levels of disciplinary structure and student support. A teacher version of the Authoritative School Climate Survey (ASCS) was administered to a statewide sample of 9099 7th- and 8th-grade teachers from 366 schools. The study used exploratory and multilevel confirmatory factor analyses (MCFA) that accounted for the nested data structure and allowed for the modeling of the factor structures at 2 levels. Multilevel confirmatory factor analyses conducted on both an exploratory (N = 4422) and a confirmatory sample (N = 4677) showed good support for the factor structures investigated. Factor correlations at 2 levels indicated that schools with greater levels of disciplinary structure and student support had higher student engagement, less teasing and bullying, and lower student aggression toward teachers. The teacher version of the ASCS can be used to assess 2 key domains of school climate and associated measures of student engagement and aggression toward peers and teachers. 

    Konold, T., Cornell, D., Huang, F., Meyer, P., Lacey, A., Nekvasil, E., Heilbrun, A., and Shukla, K. (2014). Multi-level multi-informant structure of Authoritative School Climate Survey. School Psychology Quarterly, 29, 238-255. doi: 10.1037/spq0000062

    The Authoritative School Climate Survey was designed to provide schools with a brief assessment of two key characteristics of school climate—disciplinary structure and student support—that are hypothesized to influence two important school climate outcomes—student engagement and prevalence of teasing and bullying in school. The factor structure of these four constructs was examined with exploratory and confirmatory factor analyses in a statewide sample of 39,364 students (grades 7 and 8) attending 423 schools. Notably, the analyses used a multi-level structural approach to model the nesting of students in schools for purposes of evaluating factor structure, demonstrating convergent and concurrent validity, and gauging the structural invariance of concurrent validity coefficients across gender. These findings provide schools with a core group of school climate measures guided by authoritative discipline theory.

    Lacey, A. & Cornell, D.G. (2014). School administrator assessment of bullying and state-mandated testing. Journal of School Violence. Advance online publication: doi: 10.1080/15388220.2014.971362

    This study examined the hypothesis that school administrator assessments of the prevalence of teasing and bullying (PTB) in high school are negatively associated with schoolwide performance on state-mandated testing, and that the use of evidence-based bullying prevention efforts are positively associated with test performance. School administrators from 301 high schools in the United States were surveyed on the prevalence of bullying and teasing as well as the types of bullying prevention efforts currently used in their schools. School administrator assessments of both PTB and evidence-based efforts to prevent bullying were consistently associated with the proportion of students that passed state-mandated achievement testing across 11 subject areas. School administrator assessments of schoolwide teasing and bullying, as well as their efforts to reduce it, are consistently associated with student achievement.

    Cornell, D., Gregory, A., Huang, F., & Fan, X. (2013). Perceived prevalence of teasing and bullying predicts high school dropout rates. Journal of Educational Psychology, 105, 138-149.

    This study of 281 Virginia public high schools found that the prevalence of bullying and teasing perceived by ninth grade students was predictive of dropout rates for this cohort four years later. Negative binomial regression indicated that a one SD increase in a scale measuring perceptions of bullying and teasing was associated with a 21% increase in the number of dropouts, after controlling for the effects of other predictors, including school size, student body poverty and minority composition, and performance on standardized achievement testing. The predictive value of student perceptions of bullying and teasing was comparable in magnitude to the predictive value for other commonly recognized correlates of dropout rates. These results provide new evidence that the prevalence of bullying and teasing in high school is an important factor in high school academic performance. 

    JustChildren and Cornell, D. (2013). Prevention v. punishment: Threat assessment, school suspensions, and racial disparities. Retrieved from https://www.justice4all.org/wp-content/uploads/2013/12/UVA-and-JustChildren-Report-Prevention-v.-Punishment.pdf

    Executive Summary: Racial disparities in school discipline today are troubling. Nationally, nearly one third of black male high school and middle school students undergo suspension, while only one in ten white males are suspended. In Virginia, black males are suspended at approximately twice the rate of white males in elementary, middle, and high schools. Black females are suspended at more than twice the rate of white females. There are racial disparities even when controlling for a variety of other factors, such as poverty and delinquency. Because suspension is linked to school dropout and delinquency, reducing disparities in suspension rates could help reduce school dropout and delinquency rates for all students, but especially for black males. This report presents new evidence that the implementation of Virginia Student Threat Assessment Guidelines (VSTAG) in Virginia public schools is associated with marked reductions in both short-term and long-term school suspensions. Furthermore, use of VSTAG is associated with reductions in the racial disparity in long-term suspensions. Schools using VSTAG have substantially lower rates of school suspensions, especially among black males, who tend to have the highest suspension rates. In 2013, Virginia became the first state in the country to mandate the formation of threat assessment teams in all its schools. In light of this new data, it is important for schools to take this mandate seriously. In order to reap the benefits of threat assessment, however, it must be carefully implemented and balanced with student rights, all with the goal of improving school safety and climate for everyone. In addition to the seven recommendations found at the end of this report discussing ways for schools and communities to implement threat assessment safely and fairly, we also make the following policy recommendations:1. The Virginia General Assembly should ensure that sufficient funding is available to provide school employees and law enforcement employees assigned to work in schools training in threat assessment, as well as other interventions that can help reduce suspension rates and improve student behavior. 2. The Virginia Department of Education and The Virginia Department of Criminal Justice Services should draft a model memorandum of understanding between schools and law enforcement for implementing threat assessment procedures and related efforts to maintain school safety.3. The Virginia Department of Juvenile Justice should collect data on school-based arrests, referrals to law enforcement by schools or school resource officers, and filing of delinquency petitions or criminal complaints based on conduct occurring at school. 4. The General Assembly should require that schools ensure that students who are suspended or expelled continue to make academic progress during periods of disciplinary removal.

    Lacey, A., & Cornell, D. (2013). The impact of bullying climate on schoolwide academic performance. Journal of Applied School Psychology, 29, 262-283.

    This study found that the prevalence of bullying and teasing in a high school was predictive of schoolwide performance on state-mandated achievement testing used to meet No Child Left Behind requirements. Measures of the prevalence of bullying and teasing were obtained from a statewide survey of 7,304 ninth grade students and 2,918 teachers randomly selected from 284 Virginia high schools. Hierarchical regression analyses found that the perceived prevalence bullying and teasing was predictive of schoolwide passing rates on Virginia’s Standards of Learning (SOL) tests for Algebra I, Earth Science, World History, Biology, and Geometry. 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. These results support the need for greater attention to the impact of bullying and teasing on high school student performance on high stakes testing.

    Lovegrove, P., & Cornell, D. (2013, September). Patterns of bullying and victimization associated with other problem behaviors among high school students: A conditional latent class approach. Journal of Crime and Justice. Advance online publication DOI:10.1080/0735648X.2013.832475

    Though rates of bullying are commonly found to peak in middle schools, a non-negligible amount of bullying occurs among high schools too. More information regarding patterns of bullying involvement among high school students is needed, however, as well as greater insight into the relationship high school students’ bullying involvement has with other problem behaviors. This study used latent class analysis to construct typologies of bullying involvement among over 3500 high school students from Virginia. Covariates of latent class membership were also examined in an effort to better understand the association between bullying involvement and internalizing and externalizing problem behaviors. A latent class model containing four classes was constructed, composed of a non-involved class (65%), a bullies class (12%), a victims class (16%), and a bully-victims class (8%). Externalizing problem behaviors were significantly higher among students in the bullies and bully-victims classes, while internalizing problem behaviors were higher among victims and bully-victims. Implications for the literature and for practitioners are discussed, as well as limitations and future directions.

    Mehta, S., Cornell, D., Fan, X., & Gregory, A. (2013). Bullying climate and school engagement in ninth grade students. Journal of School Health, 83, 45-52.

    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 7,058 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.

    Cornell, D., Allen, K., & Fan, X. (2012). A randomized controlled study of the Virginia Student Threat Assessment Guidelines in grades K-12. School Psychology Review, 41, 100-115. https://doi.org/10.1080/02796015.2012.12087378

    This randomized controlled study examined disciplinary outcomes for 201 students who made threats of violence at school. The students attended 40 schools randomly assigned to use the Virginia Student Threat Assessment Guidelines or follow a business-as-usual disciplinary approach in a control group. Logistic regression analyses found, after controlling for student gender, race, school level, and threat severity, that the 100 students in the threat assessment group schools were more likely to receive counseling services (odds ratio [OR] = 3.98) and a parent conference (OR = 2.57), and less likely to receive a long-term suspension (OR = 0.35) or alternative school placement (OR = 0.13) than the 101 students in the control group schools. Implementation fidelity was associated with decreased long-term suspension (OR = 0.73). These results provide strong empirical support for the use of student threat assessment in primary and secondary schools.

    Cornell, D., Klein, J., Konold, T., & Huang, F. (2012). Effects of validity screening items on adolescent survey data. Psychological Assessment 24, 21-3. doi: 10.1037/a0024824

    Two studies examined the use of validity screening items in adolescent survey data. In each study, adolescent respondents were asked whether they were telling the truth and paying attention in answering survey questions. In Study 1 (N = 7,801), the prevalence rates of student risk behaviors were significantly lower after inappropriate (“invalid”) responders were screened out of the sample. In addition, confirmatory and multi-group factor analyses demonstrated significant differences between the factor structures of school climate scales using valid versus invalid responders. In Study 2, student perceptions of school climate were correlated with teacher perceptions in 291 schools. A bootstrap resampling procedure compared the correlations obtained using valid versus invalid responding students in each school and found that valid responders had more positive views of school conditions and produced higher correlations with teacher perceptions. These findings support the value of validity screening items to improve the quality of adolescent survey data. 

    Gregory, A., Cornell, D., & Fan, X. (2012). Teacher safety and authoritative school climate in high schools. American Journal of Education, 118, 401-425.

    Most research on school climate focuses on student well-being, with less attention to the safety of school faculty. The current study examined the relationship between an authoritative school climate (characterized by high levels of student support and disciplinary structure) and both teacher reports of victimization and school records of threats against staff. Regression analyses in a statewide sample of 280 high schools showed that, both structure (as measured by student- and teacher-reported clarity of school rules) and support (as measured by teacher-reported help seeking) were associated with less teacher victimization, after controlling for school and neighborhood demographics. Support, but not structure, was a consistent predictor of school records of threats against faculty. These findings offer implications for improving the workplace for teachers and staff.

    Klein, J., Cornell, D., & Konold, T. (2012). Relationships between school climate and student risk behaviors. School Psychology Quarterly, 27, 154-169.

    This study examined whether characteristics of a positive school climate were associated with lower student risk behavior in a sample of 3,687 high school students who completed the School Climate Bullying Survey and questions about risk behavior from the Youth Risk Behavior Surveillance Survey (YRBS). Confirmatory factor analyses established reasonable fit for 20 items with three hypothesized school climate scales measuring (1) prevalence of bullying and teasing; (2) aggressive attitudes; and (3) student willingness to seek help. Structural equation modeling established the relationship of these measures with student reports of risk behavior. Multi-group analyses identified differential effects across gender and race. A positive school climate could be an important protective factor in preventing student risk behavior.

    Nekvasil, E., & Cornell, D. (2012). Student reports of peer threats of violence: Prevalence and outcomes. Journal of School Violence, 11, 357-375.

    Authorities in education and law enforcement have recommended that schools use a threat assessment approach to prevent violence, but there is relatively little research on characteristics and outcomes of threats among students. The current study examined student reports of threat experiences in a sample of 3,756 high school students. Approximately 12% of students reported being threatened at school in the past 30 days, but only 23% of threatened students regarded the threat as serious and just 26% reported the threat to school authorities. Only 9% of students who received a threat reported that it was carried out. Five reasons why students did not report threats were identified. Logistic regression analyses identified student and threat characteristics associated with threat reporting and outcome. These findings provide new information about the prevalence and nature of student threats that can inform a threat assessment approach to school violence prevention. 

    Phillips, V., & Cornell, D. (2012). Identifying victims of bullying: Use of counselor interviews to confirm peer nominations. Professional School Counseling, 15, 123-131.

    Schools often rely on anonymous self-report methods to measure bullying victimization, but these methods prevents school personnel from identifying those students who may require support. In contrast, this study employed peer nominations to identify student victims of bullying and used school counselor interviews to confirm the students’ victim status. A sample of 1,178 middle school students completed a confidential peer nomination form as part of a standard bullying survey. Students with multiple nominations were interviewed by school counselors to confirm victim status. The proportion of students confirmed as victims increased from 43% for students with two or more nominations to 90% for students with nine or more nominations.

    Baly, M., & Cornell, D. (2011). Effects of an Educational Video on the Measurement of Bullying by Self-Report, Journal of School Violence, 10, 221-238. doi: 10.1080/15388220.2011.578275

    This study of 1,283 middle school students examined the effect of an educational video designed to distinguish bullying from ordinary peer conflict. Randomly assigned classrooms of students either watched or did not watch a video prior to completing a self-report bullying survey. Compared to the control group, students who watched the video reported 32% less social bullying victimization and boys who watched the video reported 54% less physical bullying victimization and 68% less physical bullying of others. These results indicate that student self-reports could yield inflated estimates of the prevalence of bullying if students are not adequately educated about the distinction between bullying and other forms of peer conflict.

    Cornell, D., & Allen, K. (2011). Development, evaluation, and future directions of the Virginia Student Threat Assessment Guidelines. Journal of School Violence, 10, 88-106. doi: 10.1080/15388220.2010.519432

    The Virginia Student Threat Assessment Guidelines were developed in response to studies of school shootings conducted by the U.S. Federal Bureau of Investigation, the U.S. Secret Service, and U. S. Department of Education that recommended schools should adopt a threat assessment approach to prevent targeted violence. This article reviews the development and field-testing of the guidelines in a series of studies, then describes the challenges of conducting a randomized controlled trial of threat assessment. The design, measurement, and logistical challenges of conducting rigorous research on student threat assessment are discussed.

    Cornell, D., Gregory, A., & Fan, X. (2011). Reductions in long-term suspensions following adoption of the Virginia Student Threat Assessment Guidelines. Bulletin of the National Association of Secondary School Principals, 95, 175-194. https://doi.org/10.1177/0192636511415255

    This quasi-experimental study examined the adoption of the Virginia Student Threat Assessment Guidelines in 23 high schools. After training, school administrators and other staff members demonstrated substantial increases in knowledge of threat assessment principles and decreased commitment to zero tolerance approaches. Schools using the guidelines showed a 52% reduction in long-term suspensions and a 79% reduction in bullying infractions from the pretraining year to the posttraining year, in contrast to a control group of 26 schools not using the guidelines.

    Cornell, D., & Mehta, S. (2011). Counselor confirmation of middle school student self-reports of bullying victimization. Professional School Counseling, 14, 261-270.

    School counselors frequently use self-report surveys to assess bullying despite little research on their accuracy. This study raises concern that schools not rely on single self-report items to determine prevalence rates for bully victimization. In this study, counselor follow-up interviews found that only 24 (56%) of 43 middle school students who self-identified as victims of bullying could be confirmed as actual victims. Some students described peer conflicts that did not constitute bullying, mismarked the survey, or reported previous bullying that was outside the 30-day timeframe for the survey. Counselor judgments were supported by peer-nomination data and other survey responses indicative of victimization. These findings underscore the need to educate students about the definition of bullying and to use multiple sources of information in measuring the prevalence of bullying. 

    Gregory, A., Cornell, D., & Fan, X. (2011). The relationship of school structure and support to suspension rates for Black and White high school students. American Educational Research Journal, 48, 904-934. DOI: 10.3102/0002831211398531

    This study examined the relationship between structure and support in the high school climate and suspension rates in a statewide sample of 199 schools. School climate surveys completed by 5,035 ninth grade students measured characteristics of authoritative schools, defined as highly supportive, yet highly structured with academic and behavioral expectations. Multivariate analyses showed that schools low on characteristics of an authoritative school had the highest schoolwide suspension rates for Black and White students after statistically controlling for school demographics. Furthermore, schools low on both structure and support had the largest racial discipline gaps. These findings highlight the characteristics of risky settings that may not meet the developmental needs of adolescents and may contribute to disproportionate disciplinary outcomes for Black students.

    Lee, T., Cornell, D., Gregory, A., & Fan, X. (2011). High suspension schools and dropout rates for black and white students. Education and Treatment of Children, 34, 167-192.

    This study examined the association between school suspension rates and dropout rates in a statewide sample of 289 Virginia public high schools. The contribution of suspension rates on dropout rates was examined for both Black and White students, after controlling for school demographics (school racial composition, percentage of students eligible for Free and Reduced Price Meals, urbanicity), and school resources (per pupil expenditure). Because student attitudes also might influence suspension rates, the prevalence of aggressive attitudes and rejection of school rules among students were used as additional predictors. Hierarchical regression analyses using schools as the unit of analysis found that, after entering both school demographics and student attitude measures, schools with high suspension rates tended to have high dropout rates. There were comparable findings for both White and Black students, although school suspension rates were more strongly associated with White dropout rates than Black dropout rates. These findings contribute new evidence that suspension policies may have an adverse effect on student completion of high school.

  • Publications Prior to 2010

    Cornell, D., Sheras, P., Gregory, A., & Fan, X. (2009). A retrospective study of school safety conditions in high schools using the Virginia Threat Assessment Guidelines versus alternative approaches. School Psychology Quarterly, 24, 119-129. https://doi.org/10.1037/a0016182

    Threat assessment has been widely recommended as a violence prevention approach for schools, but there are few empirical studies of its use. This nonexperimental study of 280 Virginia public high schools compared 95 high schools using the Virginia threat assessment guidelines (Cornell & Sheras, 2006), 131 following other (i.e., locally developed) threat assessment procedures, and 54 not using a threat assessment approach. A survey of 9th grade students in each school obtained measures of student victimization, willingness to seek help for bullying and threats of violence, and perceptions of the school climate as caring and supportive. Students in schools using the Virginia threat assessment guidelines reported less bullying, greater willingness to seek help, and more positive perceptions of the school climate than students in either of the other 2 groups of schools. In addition, schools using the Virginia guidelines had fewer long-term suspensions than schools using other threat assessment approaches. These group differences could not be attributed to school size, minority composition or socioeconomic status of the student body, neighborhood violent crime, or the extent of security measures in the schools. Implications for threat assessment practice and research are discussed.

    Allen, K., Cornell, D., Lorek, E., & Sheras, P. (2008). Response of school personnel to student threat assessment training. School Effectiveness and School Improvement, 19, 319- 332. https://doi.org/10.1080/09243450802332184

    School safety has become an important area of concern for school improvement. This study examined the effects of staff training as means of improving school responses to student threats of violence. A multidisciplinary sample of 351 staff from 2 school divisions completed pre-post training surveys as part of a 1-day training program using the Guidelines for Responding to Student Threats of Violence (Cornell & Sheras, 2006). Analysis of pre-post surveys found large changes in staff attitudes toward school safety and violence prevention efforts. There was a substantial decrease in concerns about school homicide and increased awareness of effective violence prevention efforts. There was a drop in support for zero tolerance and profiling approaches, along with increased knowledge of threat assessment principles and concepts. These changes were sustained across school divisions serving a challenging urban population and a more affluent, suburban population. Similar effects were found across all school personnel. These findings demonstrate the viability of training staff in a student threat assessment approach.

    Strong, K., & Cornell, D. (2008). Student threat assessment in Memphis City Schools: A descriptive report. Behavioral Disorders, 34, 42-54. https://doi.org/10.1177/019874290803400104

    Threat assessment has been widely recommended as a violence prevention approach for schools but there are few reports of its implementation. Memphis City Schools adapted the Virginia threat assessment guidelines (Cornell & Sheras, 2006) for use by a centralized team serving 1 94 schools and a student population of 1 18,000. This article describes 209 student threats referred for assessment during a single school year and the resulting educational placements and disciplinary consequences. There were no reports of students carrying out any of the violent threats. These results support further examination of student threat assessment as a promising approach to dealing with student threats.

    Kaplan, S., & Cornell, D. (2005). Threats of violence by students in special education. Behavioral Disorders, 31, 107-119. https://doi.org/10.1177/019874290503100102

    We compared threats of violence made by K–12 students in special education (120 cases) or general education (136 cases) in schools that were implementing threat assessment guidelines for managing student threats of violence (Cornell, Sheras, Kaplan, McConville, Posey, Levy- Elkon, et al., 2004; Cornell & Sheras, in press). Students in special education made disproportionately more threats, as well as more severe threats, than peers in general education. Students classified as emotionally disturbed (ED) exhibited the highest threat rates. Nevertheless, use of school suspension as a disciplinary consequence for threats was consistent for students in special and general education, and few students were expelled. Our findings support the use of threat assessment to manage threats of violence by students in special education.

    Cornell, D., Sheras, P. Kaplan, S., McConville, D., Douglass, J., Elkon, A., Knight, L., Branson, C., & Cole, J. (2004). Guidelines for student threat assessment: Field-test findings. School Psychology Review, 33, 527-546.

    A demonstration project was conducted to field-test guidelines for schools to use in responding to student threats of violence. Results from 188 student threats occurring in 35 schools over the course of one school year are described. School-based teams used a decision-tree model to evaluate the seriousness of a threat and take appropriate action to reduce the threat of violence. Using threat assessment guidelines, the majority of cases (70%) were resolved quickly as transient threats. More serious cases, termed substantive threats (30%), required a more extensive evaluation and intervention plan. Follow-up interviews with school principals revealed that almost all students were able to continue in school or return to school after a brief suspension. Only 3 students were expelled, and none of the threatened acts of violence were carried out. These findings indicate that student threat assessment is a feasible, practical approach for schools that merits more extensive study.