Self Improvement

A retrospective study of school safety conditions in high schools using the Virginia threat assessment guidelines versus alternative approaches

A retrospective study of school safety conditions in high schools using the Virginia threat assessment guidelines versus alternative approaches
of 11
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  A Retrospective Study of School Safety Conditions in HighSchools Using the Virginia Threat Assessment GuidelinesVersus Alternative Approaches Dewey Cornell, Peter Sheras, Anne Gregory, and Xitao Fan University of VirginiaThreat assessment has been widely recommended as a violence prevention approach forschools, but there are few empirical studies of its use. This nonexperimental study of 280Virginiapublichighschoolscompared95highschoolsusingtheVirginiathreatassessmentguidelines (Cornell & Sheras, 2006), 131 following other (i.e., locally developed) threatassessment procedures, and 54 not using a threat assessment approach. A survey of 9thgrade students in each school obtained measures of student victimization, willingness toseek help for bullying and threats of violence, and perceptions of the school climate ascaring and supportive. Students in schools using the Virginia threat assessment guidelinesreported less bullying, greater willingness to seek help, and more positive perceptions of theschool climate than students in either of the other 2 groups of schools. In addition, schoolsusing the Virginia guidelines had fewer long-term suspensions than schools using otherthreat assessment approaches. These group differences could not be attributed to schoolsize, minority composition or socioeconomic status of the student body, neighborhoodviolent crime, or the extent of security measures in the schools. Implications for threatassessment practice and research are discussed. Keywords:  Student threat assessment, school violence, school safety, violence prevention Since the 1999 shootings at Columbine HighSchool, school administrators have been underpressure to assure the public that schools are safeand secure (Cornell, 2006). The shootings in 2005at Red Lake High School in Minnesota, in 2006 atthe Amish school in Pennsylvania, and in 2007 atVirginia Tech received worldwide attention andhave kept the issue of school safety in the fore-ground of national concerns. The purpose of thisstudy was to examine school climate conditions ina group of Virginia high schools that elected toimplement a student threat assessment programdesigned to prevent acts of violence. This inves-tigation was undertaken after a statewide surveyindicated that 95 high schools had adopted thethreat assessment guidelines developed by theUniversity of Virginia (Cornell & Sheras,2006), 54 indicated that they had no formal pro-cess, and 131 indicated that they had some othermodel. These three groups of schools were com-paredonexistingsourcesofinformationregardingstudent perceptions of school climate and levels of bullying, as well as school records of disciplinaryinfractions for aggressive behavior.Both FBI (O’Toole, 2000) and U.S. SecretService (Vossekuil, Fein, Reddy, Borum, & Dewey Cornell, Peter Sheras, Anne Gregory, Programs inClinical and School Psychology of the Curry School of Education, and Xitao Fan, Department of Educational Lead-ership and Foundations, Curry School of Education, Uni-versity of Virginia.Dewey Cornell is with the Programs in Clinical andSchool Psychology of the Curry School of Education, Uni-versity of Virginia.We thank Donna Bowman of the Virginia Department of Criminal Justice Services and Arlene Cundiff of the Vir-ginia Department of Education, and their colleagues, fortheir support of the Virginia High School Safety Study. Wealso thank our research assistants Sharmila Bandyopadhyay,Justin Collman, Megan Eliot, Francis Huang, JenniferKlein, Talisha Lee, Erica Shirley, Aisha Thompson, andFarah Williams. We thank Chris Gist for development of theGIS map that permitted us to align law enforcement crimedata with school districts. We thank James Conklin foracquiring crime data from Virginia law enforcement agen-cies and compiling the crime database. This project wassupported in part by a grant from the Office of JuvenileJustice and Delinquency Prevention of the U.S. Departmentof Justice, but the views in this article do not necessarilyreflect policies or recommendations of the funding agency.Correspondence concerning this article should be ad-dressed to Dewey Cornell, P.O. Box 400270, Curry Schoolof Education, University of Virginia 22904-4270. School Psychology Quarterly © 2009 American Psychological Association2009, Vol. 24, No. 2, 119–129 1045-3830/09/$12.00 DOI: 10.1037/a0016182 119  Modzeleski, 2002) studies remarked on the di-verse backgrounds and circumstances of stu-dents who engaged in acts of targeted violence,and identified some general characteristics seenin many, but not all, of the student perpetrators.Many of the students were victims of bullyingwho had become angry and depressed, had fam-ily relationship problems, and were negativelyinfluenced by peers. Over half displayed a pre-occupation with violence through movies orvideo games. Both law enforcement agenciesconcluded that, because these characteristicscan be found in so many students, it is notpossible to develop a profile or checklist thatcould be used to pinpoint the small number of truly violent students among them. Any check-list of warning signs would falsely identifymany students who were not dangerous.Nevertheless, the FBI and Secret Service em-phasized that almost all of these students com-municated their intentions to attack throughthreats and warnings. In most cases, the threatswere not communicated directly to the intendedvictims but to third parties such as their peers.Had these threats been reported to authoritiesand investigated, the shootings might have beenprevented; the FBI identified a series of poten-tial school shootings that were prevented be-cause students reported a threat to authoritiesthat was investigated and determined to be se-rious (O’Toole, 2000). On the basis of theseobservations, both the FBI and the Secret Ser-vice, in collaboration with the Department of Education, recommended that schools adopt a threat assessment   approach to prevent targetedacts of violence (Fein et al., 2002; O’Toole,2000). Similar recommendations were made forinstitutions of higher education following theVirginia Tech shootings (United States Depart-ment of Health and Human Services, 2007; Vir-ginia Tech Review Panel, 2007).What is threat assessment? Threat assessmentis widely used by the Secret Service to deal withpersons who threaten to attack public officials,and has evolved into a standard law enforce-ment approach to analyze a variety of dangeroussituations, such as threats of workplace vio-lence. Threat assessment is a process of evalu-ating a threat, and the circumstances surround-ing the threat, to uncover any facts or evidencethat indicate that the threat is likely to be carriedout. Student threat assessment can be distin-guished from profiling in part because the in-vestigation is triggered by the student’s ownthreatening behavior rather than some broadercombination of student characteristics.Threat assessment is ultimately concernedwith whether a student  poses  a threat, notwhether he or she has  made  a threat (O’Toole,2000; Randazzo et al., 2006). Any student canmake a threat, but relatively few will engage inthe planning and preparation necessary to carryout the threat. Threat assessment is concernedwith determining whether a student has the in-tent and means to carry out the threat and in-cludes efforts to prevent the threat from beingcarried out. Prevention efforts range from im-mediate security measures, such as notifyinglaw enforcement and warning potential victims,to the development of an intervention plan de-signed to resolve the conflict or problem thatprecipitated the threat.Although both the FBI and Secret Servicereports (Fein et al., 2002; O’Toole, 2000) madea compelling case for student threat assessment,schools had no experience with this approach,and there were many questions concerning thepractical procedures that should be followed. Inresponse, researchers at the University of Vir-ginia developed a set of guidelines for schooladministrators to use in responding to a reportedstudent threat of violence. Threat assessmentteams are trained in a 6-hr workshop that pre-pares them to use a 145-page threat assessmentmanual (Cornell & Sheras, 2006).The Virginia model of threat assessment is anapproach to violence prevention that empha-sizes early attention to problems such as bully-ing, teasing, and other forms of student conflictbefore they escalate into violent behavior.School staff members are encouraged to adopt aflexible, problem-solving approach, as distin-guished from a more punitive, zero-toleranceapproach to student misbehavior. As a result of this training, the model is intended to generatebroader changes in the nature of staff–studentinteractions around disciplinary matters and toencourage a more positive school climate inwhich students feel treated with fairness andrespect.A study of 351 school staff members whocompleted the Virginia workshop found thatparticipants became less anxious about the pos-sibility of a school homicide, more willing touse threat assessment methods to help studentsresolve conflicts, and less inclined to use a 120 CORNELL, SHERAS, GREGORY, AND FAN  zero-tolerance approach (Allen, Cornell, Lorek,& Sheras, 2008). Similar effects were found forprincipals, psychologists, counselors, socialworkers, and law enforcement officers.The Virginia guidelines include a seven-stepdecision tree. In brief, the first three steps con-stitute a triage process in which the team leader(most often a school administrator such as theprincipal or assistant principal) investigates areported threat and determines whether thethreat can be readily resolved as a  transient  threat that is not a serious threat. Examples of transient threats are jokes or statements made inanger that are expressions of feeling or figuresof speech rather than expressions of a genuineintent to harm someone.Any threat that cannot be clearly identifiedand resolved as transient is treated as a  substan-tive  threat. Substantive threats always requireprotective action to prevent the threat from be-ing carried out. The remaining four steps guidethe team through more extensive assessmentand response based on the seriousness of thethreat. In the most serious cases, the team con-ducts a safety evaluation that includes both alaw enforcement investigation and a mentalhealth assessment of the student. The culmina-tion of the threat assessment is the developmentof a safety plan that is designed to address theproblem or conflict underlying the threat andprevent the act of violence from taking place.For both transient and substantive threats, thereis an emphasis on helping students resolve con-flicts and minimizing the use of zero-tolerancesuspensions as a disciplinary response.The Virginia threat assessment guidelineswere field tested in 35 public schools, encom-passing an enrollment of more than 16,000 stu-dents in Grades K–12 (Cornell et al., 2004).School-based teams evaluated 188 studentthreats that involved threats to hit, stab, shoot,or harm someone in some other way. Most of the threats (70%) were resolved as transientthreats, and the remaining 30% were substantivethreats that required more extensive assessmentandprotectiveaction.Thethreatassessmentteamsplaced special emphasis on understanding thecontext and meaning of the threat and develop-ing a plan to address the underlying conflict orproblem that stimulated the student to resort tothreatening behavior. Use of this problem-solving approach meant that relatively fewstudents received long-term suspensions or ex-pulsions from school. Only 3 students wereexpelled from school, although half of the stu-dents ( n  94) received short-term suspensions(typically 1–3 days). Notably, follow-up inter-views with the school principals found no casesin which the threats were carried out.A second study examined the Virginia threatassessment model when used by a centralizedteam responding to 209 serious threat cases inMemphis City Schools (Strong & Cornell,2008). There were 60 (29%) threats to hit orbeat up someone, 48 (23%) threats to cut orstab, 32 (15%) threats to shoot, 30 (14%) threatsto kill, 14 (7%) sexual threats, and 25 (12%)other threats (such as to blow up or burn downthe school). This study found that all of thestudent threats were resolved without any de-tected act of violence. Almost all students wereable to return to their school or an alternativeschool placement, with only five students re-ceiving long-term suspensions without schoolservices. Plans to assist each student includedmodifications to special education plans, theprovision of academic and behavioral supportservices, and referrals to community-basedmental health services. After the threat assess-ment, the number of disciplinary office referralsfor these students declined by approximately55% through the remainder of the school year.The most notable limitation to previous stud-ies of the Virginia threat assessment model isthe absence of a comparison group. To addressthis need, the present study examined the use of the Virginia threat assessment model in thestatewide population of Virginia high schools.The 95 high schools using the Virginia modelwere compared with 131 schools using a locallydeveloped threat assessment model and 54schools not using a threat assessment approach.This was a retrospective comparison conductedafter the school principals had responded to aquestion on an annual school safety audit surveyabout their approach to threat assessment.We expected that schools using the Virginiamodel of threat assessment would create a morepositive and supportive school climate that en-couraged students to come forward to obtainhelp in response to bullying and threats of vio-lence, and that this in turn would give staff moreopportunities to prevent or reduce student bul-lying and other forms of victimization. We ex-pected that schools using the Virginia model toresolve student conflicts would be less likely to 121THREAT ASSESSMENT GUIDELINES  use school exclusion as a response to disciplin-ary infractions.Data on student victimization and perceptionsof school climate were available from the VirginiaHigh School Safety Study (Cornell & Gregory,2008), a statewide examination of school climateand safety conditions in Virginia public highschools using data collected from school princi-pals, students, teachers, and school records. Thepurpose of the study was to identify school safetypractices that were associated with more positiveschool climates and lower levels of crime andviolence. Most relevant to the present study, theVirginia High School Safety Study included astatewide survey of ninth grade students. Ninthgrade students were surveyed because the firstyearofhighschoolisconsideredapivotalyearforstudent adjustment and achievement (Donegan,2008), ninth grade students in Virginia have anespecially high rate of discipline violations (Vir-giniaDepartmentofEducation,2007),andnation-ally, ninth grade students experience a high rate of bully victimization (Nansel et al., 2001), probablybecause they are youngest students in the school.This study did not collect case data on studentthreats, so schools were compared on the basis of more general outcomes that could be expectedfromtheadoptionofathreatassessmentapproach.Consequently, we hypothesized that schoolsusing the Virginia model would have lowerrates of long-term suspensions and fewer disci-plinary violations involving aggressive behav-ior. We further hypothesized that there wouldbe less student bullying and victimization, andthat students would have a positive view of theschool climate if the school adopted a problem-solving approach, rather than the more punitive,zero-tolerance approach that is widely adoptedin Virginia schools. Finally, we hypothesizedthat students would be more willing to seek helpfrom school staff for bullying and other threatsof violence, and that they would have a morepositive perception of school staff as treatingthem with fairness and respect. Method ParticipantsSchools All 314 Virginia high schools were eligiblefor inclusion in the Virginia High School SafetyStudy, which was the source of data for thisreport. Virginia law requires every publicschool principal to complete an annual onlineschool safety audit. The principal survey for the2006–2007 school year asked whether theyused “a formal threat assessment process torespond to student threats of violence.” In re-sponse, 95 principals checked the answer “Yes,we follow the guidelines developed by the Uni-versity of Virginia (UVA),” 54 indicated thatthey had no formal process, and 131 indicatedthat they had some other process. In response toa follow-up question about the source of theirguidelines, these principals wrote that they weredeveloped by some combination of in-houseadministrative staff (52 schools), by district-level staff (48 schools), or a combination of school staff and local professionals in law en-forcement or mental health (6 schools). Twoprincipals reported that they did not know thesource of their guidelines, 1 school reported useof a private consultant, and 1 reported that theyused state department of education guidelines(although such guidelines do not exist). Theremaining 34 schools did not provide a responseand could not be included in the study.The 280 participating schools ranged in sizefrom 33 to 2,881 students, with an averageof 1,199 students. All 280 schools participatedin the Virginia High School Safety Study (de-scribed below). There were 50 urban, 110 sub-urban, and 120 rural schools. The percentage of minority students in the schools ranged from0% to 100%, with an average of 34% ( SD   26). The percentage of students eligible for re-duced price meals ranged from 0% to 100%,with an average of 31% ( SD  16). The numberof school resource officers at the schools rangedfrom none to three, including 36 schools withno officer, 232 with one officer, 10 with twoofficers, and 2 with three officers. Students As part of the Virginia High School SafetyStudy (Cornell & Gregory, 2008), school prin-cipals selected approximately 25 ninth gradestudents per school by matching a series of random numbers to alphabetized student rolls.(Schools with fewer than 25 ninth grade stu-dents selected all available ninth grade stu-dents.) Principals were instructed to send a stan-dard letter to parents explaining that their son or 122 CORNELL, SHERAS, GREGORY, AND FAN  daughter had been chosen to complete an anon-ymous online survey as part of the state’s schoolsafety audit program and advising them to con-tact the school if they did not wish their child toparticipate. Students who were unwilling or un-available to complete the survey were replacedwith the next available student on the list.Principals reported that approximately 27%of the students initially identified by the sam-pling procedure did not participate in the study.The reasons for nonparticipation included stu-dent declined to participate (16% of those whodid not participate), parent declined (6%), stu-dent absent due to illness (32%), student sus-pended from school (5%), student moved ortransferred (7%), student language barrier (3%),or some other reason (this could range from asevere disability to attending a field trip; 30%).The student participants consisted of 7,318ninth grade students (49% female) with an av-erage age of 14.8 years and a range of 12 to 17years (87% were ages 14 or 15). The self-reported racial/ethnic distribution of the samplewas 63% White/Caucasian, 23% Black/AfricanAmerican, 5% Latino/Hispanic, 3% AsianAmerican, 1% American Indian, and 5% other.  Measures Disciplinary Records High school principals in Virginia reportstudent suspensions and other disciplinary ac-tions to the Virginia Department of Educationusing a standard set of reporting conventionsand 113 categories of disciplinary infractions.State records for the 2006–2007 school yearprovided the number of long-term suspen-sions (  5 days) and short-term suspensions(  5 days) for each high school. The categoryof long-term suspensions also included expul-sions because there were too few expulsions(  Mdn    0) to justify separate analyses. Inaddition, the numbers of disciplinary referralsfor aggressive behavior (all forms of assaultand physical altercation, fighting, bullying,possession of a weapon) were summed into atotal score. On the school safety audit survey,school principals reported the number of schoolresourceofficersemployedattheschoolonadailybasis.  Neighborhood Violent Crime To measure the extent of violent crime in theneighborhoods comprising the high school at-tendance zones, we mapped annual records ob-tained from the Virginia Department of StatePolice and local law enforcement agencies ontoschool attendance zones. The total numbers of violent crimes using standard FBI definitions of violent crime were identified. Crimes occurringat school were not included in the count. Student Survey Ninth grade students completed a school cli-mate survey as part of the Virginia High SchoolSafety Study in the spring of 2007. The surveywas completed anonymously online at com-puter stations in classrooms. Student responsesat each school were aggregated into school-level scores.Student perceptions of school security weremeasured by a nine-item Security Measures In-dex derived from the School Crime Supplementto the National Crime Victimization Survey(National Center for Education Statistics,2005). Students were asked whether theirschool had each of nine security measures inplace (responding  yes ,  no ,  don’t know ), such as“security guards or assigned police officers,”“metal detectors,” and “one or more securitycameras to monitor the school.” The averagenumber of security measures identified by thestudents at each school was used as an index of school security efforts.The survey included a Victimization Indexfrom the Effective School Battery (Gottfredson,1999). Students were asked (true or false)whether each of seven forms of criminal vic-timization had happened to them in school.Items ranged from theft of personal property tobeing physically attacked. Internal consistency(Cronbach’s alpha) of this index was .68 in thesample for the Virginia High School SafetyStudy.Two measures of bullying were included inthe high school survey. Both measures weretaken from the School Climate Bullying Survey(Cornell & Sheras, 2003) and have been used inother studies of bullying (Branson & Cornell, inpress; Cornell & Brockenbrough, 2004; Thun-fors & Cornell, 2008; Williams & Cornell,2006). The Bullying Climate Scale consisted of  123THREAT ASSESSMENT GUIDELINES
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks