PROTECTING AMERICA’S SCHOOLS
A U.S. SECRET SERVICE ANALYSIS
OF TARGETED SCHOOL VIOLENCE
U.S. Department of Homeland Security
UNITED STATES SECRET SERVICE
National Threat Assessment Center
2019
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
National Threat Assessment Center
U.S. Secret Service
U.S. Department of Homeland Security
November 2019
This publication is in the public domain. Authorization to copy and distribute this publication in whole or in part is granted. However, the
U.S. Secret Service star insignia may not be otherwise reproduced or used in any other manner without advance written permission
from the agency. While permission to reprint this publication is not necessary, when quoting, paraphrasing, or otherwise referring to this
report, the citation should be: National Threat Assessment Center. (2019). Protecting America’s Schools: A U.S. Secret Service Analysis of
Targeted School Violence. U.S. Secret Service, Department of Homeland Security.
This report was authored by the staff of the
U.S. Secret Service National Threat Assessment Center (NTAC)
Lina Alathari, Ph.D.
Chief
Front Cover:
FEBRUARY 15-19, 2018
Flags flown at half-staff in honor of the victims of the shooting at
Marjory Stoneman Douglas High School in Parkland, Florida
Diana Drysdale, M.A.
Supervisory Social Science Research Specialist
Ashley Blair, M.A.
Social Science Research Specialist
Arna Carlock, Ph.D.
Social Science Research Specialist
Aaron Cotkin, Ph.D.
Social Science Research Specialist
Brianna Johnston, M.A.
Social Science Research Specialist
Christina Foley, M.S.W.
Assistant to the Special Agent in Charge
Peter Langman, Ph.D.
Psychologist and Author
Eric B. Elbogen, Ph.D.
Professor of Psychiatry and
Behavioral Sciences, Duke
University School of Medicine
Wade Buckland
Senior Scientist,
Human Resources
Research Organization
Special thanks to the following for their contributions to the project:
Steven Driscoll, M.Ed.
Lead Social Science Research Specialist
David Mauldin, M.S.W.
Social Science Research Specialist
Jeffrey McGarry, M.A.
Social Science Research Specialist
Jessica Nemet, M.A.
Social Science Research Specialist
Natalie Vineyard, M.S.
Social Science Research Specialist
John Bullwinkel, M.A.
Assistant to the Special Agent in Charge
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
The U.S. Secret Service has a longstanding tradition of conducting threat assessments as part of its mandate to ensure
the safety of this Nation’s highest elected ofcials. Our National Threat Assessment Center (NTAC) is dedicated to
expanding the eld of violence prevention by closely examining the targeted violence that affects communities across the
United States. As part of this mission, NTAC has maintained a particular focus on the prevention of targeted school
violence. For 20 years, the Center has studied these tragedies, and the following report, titled Protecting America’s
Schools: A U.S. Secret Service Analysis of Targeted School Violence, supports past Secret Service research ndings that
indicate targeted school violence is preventable.
While communities can advance many school safety measures on their own, our experience tells us that keeping schools
safe requires a team effort and the combined resources of the federal, state, and local governments; school boards; law
enforcement; and the public. With this study, the Secret Service provides an unprecedented base of facts about school
violence, as well as an updated methodology and practical guidelines for prevention. We encourage all of our public safety
partners and education partners to review the information within, and to use it to guide the best practices for maintaining
a safe learning environment for all children.
For 150 years, the men and women of the Secret Service have carried out our no-fail mission to protect the Nation’s
leaders and nancial infrastructure. Our relationships across the federal, state, and local levels have been instrumental
to our success. The agency is committed to carrying on this collaborative approach to better protect our children and our
schools. We are proud to release this report.
The U.S. Secret Service’s National Threat Assessment Center (NTAC) was created in 1998 to provide guidance on threat assessment both within the U.S.
Secret Service and to others with criminal justice and public safety responsibilities. Through the Presidential Threat Protection Act of 2000, Congress
formally authorized NTAC to conduct research on threat assessment and various types of targeted violence; provide training on threat assessment and
targeted violence; facilitate information-sharing among agencies with protective and/or public safety responsibilities; provide case consultation on
individual threat assessment investigations and for agencies building threat assessment units; and develop programs to promote the standardization of
federal, state, and local threat assessment processes and investigations.
James M. Murray
Director
MESSAGE FROM THE DIRECTOR
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
Ensuring the safety of children at school is a responsibility that belongs to everyone, including law enforcement, school
staff, mental health practitioners, government ofcials, and members of the general public. To aid in these efforts, the U.S.
Secret Service National Threat Assessment Center (NTAC) studied 41 incidents of targeted school violence that occurred
at K-12 schools in the United States from 2008 to 2017. This report builds on 20 years of NTAC research and guidance
in the eld of threat assessment by offering an in-depth analysis of the motives, behaviors, and situational factors of the
attackers, as well as the tactics, resolutions, and other operationally-relevant details of the attacks.
The analysis suggests that many of these tragedies could have been prevented, and supports the importance of schools
establishing comprehensive targeted violence prevention programs as recommended by the Secret Service in Enhancing
School Safety Using a Threat Assessment Model: An Operational Guide for Preventing Targeted School Violence.
1
This
approach is intended to identify students of concern, assess their risk for engaging in violence or other harmful activities,
and implement intervention strategies to manage that risk. The threshold for intervention should be low, so that schools
can identify students in distress before their behavior escalates to the level of eliciting concerns about safety.
Because most of these attacks ended very quickly, law enforcement rarely had the opportunity to intervene before serious
harm was caused to students or staff. Additionally, many of the schools that experienced these tragedies had implemented
physical security measures (e.g., cameras, school resource ofcers, lockdown procedures). Prevention is key.
Some of the key ndings from this study, and their implications for informing school violence prevention efforts, include:
• There is no profile of a student attacker, nor is there a profile for the type of school that has been targeted:
Attackers varied in age, gender, race, grade level, academic performance, and social characteristics. Similarly,
there was no identied prole of the type of school impacted by targeted violence, as schools varied in size,
location, and student-teacher ratios. Rather than focusing on a set of traits or characteristics, a threat assessment
process should focus on gathering relevant information about a student’s behaviors, situational factors, and
circumstances to assess the risk of violence or other harmful outcomes.
• Attackers usually had multiple motives, the most common involving a grievance with classmates: In addition to
grievances with classmates, attackers were also motivated by grievances involving school staff, romantic
relationships, or other personal issues. Other motives included a desire to kill, suicide, and seeking fame or
notoriety. Discovering a student’s motive for engaging in concerning behavior is critical to assessing the
student’s risk of engaging in violence and identifying appropriate interventions to change behavior and
manage risk.
• Most attackers used firearms, and firearms were most often acquired from the home: Many of the attackers
were able to access rearms from the home of their parents or another close relative. While many of the rearms
were unsecured, in several cases the attackers were able to gain access to rearms that were secured in a locked
gun safe or case. It should be further noted, however, that some attackers used knives instead of rearms to
perpetrate their attacks. Therefore, a threat assessment should explore if a student has access to any weapons,
with a particular focus on weapons access at home. Schools, parents, and law enforcement must work together
rapidly to restrict access to weapons in those cases when students pose a risk of harm to themselves or others.
• Most attackers had experienced psychological, behavioral, or developmental symptoms: The observable mental
health symptoms displayed by attackers prior to their attacks were divided into three main categories:
psychological (e.g., depressive symptoms or suicidal ideation), behavioral (e.g., deance/misconduct or symptoms
of ADD/ADHD), and neurological/developmental (e.g., developmental delays or cognitive decits). The fact that half
EXECUTIVE SUMMARY
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
of the attackers had received one or more mental health services prior to their attack indicates that mental health
evaluations and treatments should be considered a component of a multidisciplinary threat assessment, but not
a replacement. Mental health professionals should be included in a collaborative threat assessment process that
also involves teachers, administrators, and law enforcement.
• Half of the attackers had interests in violent topics: Violent interests, without an appropriate explanation, are
concerning, which means schools should not hesitate to initiate further information-gathering, assessment, and
management of the student’s behavior. For example, a student who is preoccupied or xated on topics like the
Columbine shooting or Hitler, as was noted in the backgrounds of several of the attackers in this study, may be the
focus of a school threat assessment to determine how such an interest originated and if the interest negatively
impacts the student’s thinking and behavior.
• All attackers experienced social stressors involving their relationships with peers and/or romantic partners:
Attackers experienced stressors in various areas of their lives, with nearly all experiencing at least one in the six
months prior to their attack, and half within two days of the attack. In addition to social stressors, other stressors
experienced by many of the attackers were related to families and conicts in the home, academic or disciplinary
actions, or other personal issues. All school personnel should be trained to recognize signs of a student in crisis.
Additional training should focus on crisis intervention, teaching students skills to manage emotions and resolve
conicts, and suicide prevention.
• Nearly every attacker experienced negative home life factors: The negative home life factors experienced by
the attackers included parental separation or divorce, drug use or criminal charges among family members, and
domestic abuse. While none of the factors included here should be viewed as predictors that a student will be
violent, past research has identied an association between many of these types of factors and a range of negative
outcomes for children.
• Most attackers were victims of bullying, which was often observed by others: Most of the attackers were
bullied by their classmates, and for over half of the attackers the bullying appeared to be of a persistent pattern
which lasted for weeks, months, or years. It is critical that schools implement comprehensive programs designed to
promote safe and positive school climates, where students feel empowered to report bullying when they witness it
or are victims of it, and where school ofcials and other authorities act to intervene.
• Most attackers had a history of school disciplinary actions, and many had prior contact with law enforcement:
Most attackers had a history of receiving school disciplinary actions resulting from a broad range of inappropriate
behavior. The most serious of those actions included the attacker being suspended, expelled, or having law
enforcement interactions as a result of their behavior at school. An important point for school staff to consider is
that punitive measures are not preventative. If a student elicits concern or poses a risk of harm to self or others,
removing the student from the school may not always be the safest option. To help in making the determination
regarding appropriate discipline, schools should employ disciplinary practices that ensure fairness, transparency
with the student and family, and appropriate follow-up.
• All attackers exhibited concerning behaviors. Most elicited concern from others, and most communicated their
intent to attack: The behaviors that elicited concern ranged from a constellation of lower-level concerns to
objectively concerning or prohibited behaviors. Most of the attackers communicated a prior threat to their target or
communicated their intentions to carry out an attack. In many cases, someone observed a threatening
communication or behavior but did not act, either out of fear, not believing the attacker, misjudging the immediacy or
location, or believing they had dissuaded the attacker. Students, school personnel, and family members should be
encouraged to report troubling or concerning behaviors to ensure that those in positions of authority can intervene.
A multidisciplinary threat assessment team, in conjunction with the appropriate policies, tools, and training, is the best
practice for preventing future tragedies. A thorough review of the ndings contained in this report should make clear that
tangible steps can be taken to reduce the likelihood that any student would cause harm, or be harmed, at school.
TABLE OF CONTENTS
INTRODUCTION ................................................................................................................................1
PART I: OVERVIEW OF THE ATTACKS ................................................................................................ 7
School & Community Information .......................................................................................................7
School Security & Prevention Measures .............................................................................................8
Weapons Used ................................................................................................................................. 9
Timing & Location ............................................................................................................................9
Response Times ............................................................................................................................11
Resolution & Harm .........................................................................................................................12
Judicial Outcomes ..........................................................................................................................13
PART II: THE ATTACKERS ...............................................................................................................14
Motives .........................................................................................................................................15
Targeting........................................................................................................................................18
Planning ........................................................................................................................................20
Firearm Acquisition .........................................................................................................................22
Law Enforcement Contact ...............................................................................................................23
Psychological, Behavioral, & Developmental Health ...........................................................................24
Substance Use & Abuse .................................................................................................................27
Weapons Use & Violence ................................................................................................................27
Home Life Factors ..........................................................................................................................29
Stressors ......................................................................................................................................31
Bullying .........................................................................................................................................33
School Interactions & Relationships ................................................................................................37
Academic Performance & Extracurricular Activities ............................................................................37
Disciplinary History.........................................................................................................................38
Threat Assessment ........................................................................................................................42
Concerning Behaviors .....................................................................................................................43
PART III: IMPLICATIONS.................................................................................................................49
APPENDIX A: STATISTICAL ANALYSES ...............................................................................................55
APPENDIX B: COMPARISON TO THE SAFE SCHOOL INITIATIVE ............................................................57
PROTECTING AMERICA’S SCHOOLS
A U.S. SECRET SERVICE ANALYSIS
OF TARGETED SCHOOL VIOLENCE
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
1
For the past 20 years, the U.S. Secret Service National Threat Assessment Center (NTAC) has been conducting a
unique blend of operationally relevant and behavior-based research on the prevention of targeted violence in various
contexts, including attacks targeting public ofcials and public gures, government facilities, workplaces, public
spaces, K-12 schools, and institutions of higher education. Targeted violence is a term coined by the Secret Service to
refer to any incident of violence where an attacker selects a particular target prior to an attack. The Presidential Threat
Protection Act of 2000 congressionally authorized NTAC to conduct research, training, consultation, and
information-sharing on the prevention of targeted violence, and to provide guidance to law enforcement, government
agencies, schools, and other public safety and security professionals.
Threat assessment is the best practice for preventing incidents of targeted violence. It is an investigative approach
pioneered by the Secret Service, originally developed as a means to prevent assassinations. A threat assessment,
when conducted by the Secret Service, involves identifying
individuals who have a concerning or threatening interest
in the president of the United States or another protected
person, conducting an investigation to assess whether
that individual poses a risk of violence or other unwanted
outcome, and then taking steps to manage that risk.
These cases receive the highest priority of all Secret
Service investigations, and the agency considers these
investigations to be as important as the physical security
measures it employs.
The Secret Service’s threat assessment model has since
been adapted to prevent other acts of targeted violence
impacting communities across the United States. These
attacks have a profound and devastating impact on those
directly affected and the Nation as a whole, none more so
than attacks at K-12 schools.
2
Following the tragedy at
Columbine High School in 1999, the Secret Service
partnered with the U.S. Department of Education on a
study that examined 37 incidents of targeted school
violence that occurred from 1974 to 2000. The Safe
School Initiative focused on gathering and analyzing
information about the thinking and behavior displayed by
the students who committed these violent acts.
3
The
publication of the nal report and accompanying guide
provided schools and law enforcement with a framework
for how to identify, assess, and manage students who
display such threatening or concerning behavior.
Key Findings from the Safe School Initiative
(2002)
• Incidents of targeted violence at school rarely were
sudden or impulsive acts.
• Prior to most incidents, other people knew about the
attacker’s idea and/or plan to attack.
• Most attackers did not threaten their targets directly
prior to advancing the attack.
• There is no accurate or useful “profile” of students
who engaged in targeted school violence.
• Most attackers engaged in some behavior prior to the
incident that caused others concern or indicated a
need for help.
• Most attackers had difficulty coping with significant
losses or personal failures. Moreover, many had
considered or attempted suicide.
• Many attackers felt bullied, persecuted, or injured by
others prior to the attack.
• Most attackers had access to and had used weapons
prior to the attack.
• Most attackers demonstrated some interest in
violence, through movies, video games, books, or
other media.
• Most attackers had no history of prior violent or
criminal behavior.
• In many cases, other students were involved in some
capacity.
• Despite prompt law enforcement responses, most
attacks were stopped by means other than law
enforcement intervention.
INTRODUCTION
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
2
Since the release of the study in 2002, the ndings of the Safe School Initiative have informed the targeted violence
prevention efforts of schools and law enforcement across the country. To date, NTAC has delivered over 500 trainings
on K-12 school threat assessment to over 100,000 school personnel, law enforcement, mental health professionals,
and others with school safety responsibilities. NTAC has further consulted with individual schools and school districts,
as well as county and state governments, as they developed threat assessment protocols geared toward proactively
preventing these tragedies.
There is more work to be done.
In 2018, two incidents of targeted school violence alone resulted in the tragic loss of 27 students and staff, and
serious injuries to 30 more. The February 14, 2018 attack at Marjory Stoneman Douglas High School in Parkland, FL
and the May 18, 2018 attack at Santa Fe High School in Santa Fe, TX prompted school personnel, law enforcement,
government agencies, and others to identify strategies for preventing future attacks. To support these efforts, NTAC
initiated programs to provide updated research and guidance on threat assessment and the prevention of targeted
school violence.
The rst phase involved the creation and distribution of an operational
guide titled Enhancing School Safety Using a Threat Assessment
Model: An Operational Guide for Preventing Targeted School
Violence. Released in July 2018, the guide outlines actionable steps
schools can take to create multidisciplinary threat assessment
teams, establish central reporting mechanisms, identify student
behaviors of concern, dene the threshold for law enforcement
intervention, promote safe school climates, and identify
intervention and management strategies for decreasing the risk
of a targeted attack. The guide is available on the Secret
Service public website and was distributed to 40,000 public
school districts and private schools across
the country.
The second phase was to conduct a new research study
expanding on the Secret Service’s previous work in
studying targeted school violence. The report that follows is
the culmination of these efforts and represents the most
in-depth analysis of targeted school violence conducted
in decades. Protecting America’s Schools: A U.S. Secret
Service Analysis of Targeted School Violence examines
41 attacks against K-12 schools in the United States
from 2008 to 2017. The report examines the backgrounds and
behaviors of the attackers, in order to inform the best practices of
multidisciplinary school threat assessment programs nationwide.
ENHANCING SCHOOL SAFETY
USING A THREAT ASSESSMENT MODEL
An Operational Guide for
Preventing Targeted School Violence
National Threat Assessment Center
July 2018
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
Methodology
Incident Identication
NTAC researchers initiated a systematic review of relevant databases, publications, and public reports. The review
sought to identify incidents of targeted school violence that occurred in the United States from 2008 to 2017. For
this study, an incident of targeted school violence was dened as any incident in which (i) a current or recently former
4
K-12 school student (ii) purposefully used a weapon (iii) to cause physical injury to, or the death of, at least one other
student and/or school employee (iv) in or on the immediate property of the school (v) while targeting in advance one
or more specic and/or random student(s) and/or employee(s).
Certain exclusions were used in order to focus this project on incidents of targeted school violence. This study does
not include attacks where a perpetrator could not be identied, or incidents related to gang violence, drug violence, or
other incidents with a strong suggestion of a separate criminal nexus. Similarly, it does not include violence from the
surrounding community that spilled onto school property by happenstance. It also excludes spontaneous acts, such
as those that were the immediate result of an unplanned ght or other sudden confrontation.
Because this project was initiated in early 2018, incident identication and collection were limited to cases that
occurred through 2017, thereby allowing researchers to gather the most comprehensive case information for each
incident. For this reason, the tragedies that have occurred in 2018 and 2019 were not included in the analysis.
Based on the stated criteria, researchers identied 41 incidents of targeted school violence, perpetrated by 41 current
or recently former students, from January 2008 through December 2017.
Case Information
Researchers largely obtained information for the 41 identied incidents from primary source materials related to the
incident. Secret Service eld ofces across the United States worked with their local law enforcement partners to
acquire investigative case les for 36 of the 41 incidents. These les included police investigative records, publicly
available court records, and other publicly available information. These source materials may have also contained
school records and mental health records. Obtained records included interviews with the attackers, interviews with
witnesses and people who knew the attackers, school transcripts and disciplinary histories, social media screen
captures, data from the searches of phones and computers used by the attackers, the results of searches of the
attackers’ residences, personal journals and other writings, and court records containing the results of mental health
evaluations both before and after the attacks. All case examples used in the drafting of this report were vetted
through the agencies who provided the investigative les. Case analysis was further supplemented through a
rigorous, structured review of open source information, including news articles and reports from government and
private agencies. NTAC staff considered all available qualitative information to develop data relevant to threat
assessment and prevention factors for each case. Some data were analyzed and are reported here for all 41 cases,
including information on how the attacks were executed, school information, and demographic information about the
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
3
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019
attackers. Information on the behavioral histories of the attackers, however, was only analyzed for 35 of the 41 cases
due to the limited information available on the backgrounds of 6 of the attackers.
Multiple layers of review were applied to every stage of the project to ensure accuracy and reliability of
reported ndings.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
4
5
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
6
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PART I: OVERVIEW OF THE ATTACKS
School & Community Information
SCHOOL TYPES: All but two of the attacks occurred in public schools (n = 39, 95%). Nearly three-quarters of the
attacks were carried out at high schools (n = 30, 73%), which are dened in this report as any school that reaches
12th grade. Nine attacks took place at middle schools (22%), and one attack occurred at an elementary school (2%).
One additional attack occurred at a K-11
school (2%).
COMMUNITIES: Approximately one-third (n = 14,
34%) of the attacks occurred in schools located in
suburban communities, while 11 attacks occurred
in cities (27%). Ten incidents were carried out in
rural schools (24%), and the fewest attacks took
place in small towns (n = 6, 15%).
5
SCHOOL SIZE: Over half (n = 24, 59%) of the attacks took place at
schools with fewer than 1,000 students, and 17 (41%) occurred at
schools with 1,000 students or more. Over one-third of the attacks took
place at schools with between 500 and 999 students (n = 15, 37%).
CLASS SIZE: Teacher-to-pupil ratios also varied among the impacted
schools, with eight of the schools (20%) at the national average of 1:15
to 1:16. One-third (n = 13, 32%) had lower ratios than the average, and
20 schools (49%) had higher ratios. The highest ratio was 1 teacher for
every 26 students.
7
* National average
Percentages may not total 100 due to rounding.
Teacher-to-Pupil Ratios
1:9 to 1:14
1:15 to 1:16*
6
1:20 to 1:26
1:17 to 1:19
32%13
20%
27%
8
11
22%9
Elementary School 2%
Other School 2%
SCHOOL TYPE
Middle
School
22%
High
School
73%
7
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019
Student Enrollment
<500
1000 - 1499
500 - 999
1500+
22%
20%
9
8
37%
22%
15
9
Percentages may not total 100 due to rounding.
Percentages may not total 100 due to rounding.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
8
School Security & Prevention Measures
Most of the schools (n = 33, 80%) in this study implemented some type of physical security measure, and two-thirds
(n = 27, 66%) had full- or part-time school resource officers (SROs) on campus.
PHYSICAL SECURITY MEASURES: The most frequently used physical security measure was school lockdown
procedures, present in at least 28 (68%) of the schools where an attack took place. Fourteen (34%) schools had
security cameras, which were located either inside or outside of the
schools. Seven (17%) had alert systems, which were used to notify
members of the school community of emergencies via automated
text messages or phone calls. Three (7%) schools had
magnetometers, but they were not regularly used. Six (15%) schools
employed private security guards, who were usually unarmed.
SCHOOL RESOURCE OFFICERS: At the time of the attacks, nearly
half of the schools (n = 19, 46%) had one or more full-time SROs.
An additional eight schools (20%) had an SRO who operated at the
school part-time because they were assigned to multiple schools
within the district.
REPORTING TOOLS: Only seven (17%) schools had any type of
system in place to notify school staff or administrators of
threatening or concerning student behaviors before an attack.
These systems usually involved a phone number, email address, or
a paper referral through which the school could be contacted. At the
time of the incidents, few states had
implemented comprehensive statewide reporting
programs, like Safe2Tell™ Colorado.
ASSESSMENT AND INTERVENTION PROTOCOLS:
Nine (22%) schools had some type of program
involving employees who were assigned to
assess unwanted or potentially harmful student
behavior. Some of these schools had developed
basic protocols for assessing and responding
to reports of a student threat. In others, school
staff created more formal threat assessment
teams, but the participation, training, and
protocols of these teams varied. The Secret
Service recommends that schools implement
multidisciplinary school threat
Physical Security and
Threat Assessment
When the Safe School Initiative was published in 2002,
it stated that “the Secret Service considers threat
assessment to be as important to preventing targeted
violence as the physical measures it employs." The
Secret Service protects our Nation’s highest elected
ofcials using physical security measures in
conjunction with, and complemented by, a threat
assessment approach designed to proactively
intervene with those individuals who intend to cause
harm. This complementary approach to protection
recognizes that either of the approaches, alone, would
not constitute the most effective means of preventing
an attack.
Safe2Tell
TM
Colorado
Safe2Tell is a statewide, anonymous reporting tool, which accepts tips 24/7
regarding any concern of safety to self or others. Every tip is evaluated by
Safe2Tell analysts at the Colorado Information Analysis Center (CIAC), then
disseminated to schools and local law enforcement as appropriate. Built-in
accountability ensures that every tip is responded to before a case can
be closed.
During the 2018-2019 school year, 19,861 tips were received by Safe2Tell.
The most common tips received related to student suicide, drug use,
and bullying.
For states that would like to implement this type of reporting program,
Safe2Tell provides information and resources at:
www.safe2tell.org
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
9
assessment teams in accordance with Enhancing School Safety Using a Threat Assessment Model: An Operational
Guide for Preventing Targeted School Violence. This framework is intended to help schools develop the capacity to
identify, assess, and manage students who display concerning or threatening behaviors. This process should
complement the physical security measures that a school determines are appropriate for its community.
Weapons Used
Most of the attackers (n = 25, 61%) used rearms, which included handguns, ries, and
shotguns. In total, attackers used 18 handguns and 9 long guns during the attacks,
with two attackers using multiple rearms. The remainder (n = 16, 39%) used bladed
weapons, which most frequently included pocket or folding knives, followed by butcher
or kitchen knives, and hunting knives. In one instance, the attacker used a World War II
bayonet. Three of the attackers used a combination of weapons to cause harm or dam-
age, including one who used a knife and a claw hammer, another who used a knife and
a bo staff, and another who used a rearm and a Molotov cocktail.
Several of the attackers brought other weapons to the attack that they did not
ultimately use. Some brought knives in addition to the rearm or knife that was used,
while others brought items ranging from a wrench wrapped in a bandana to homemade explosives. For example, one
attacker who used a knife in his attack also brought ve other knives, a blowtorch, three containers of ammable
liquid, and recrackers.
Timing & Location
TIMING OF THE ATTACKS: The 41 attacks occurred with
varying frequency from year to year and did not appear to
be steadily increasing or decreasing. While these
incidents comprise only some of the acts of violence
that occur in K-12 schools each year, data from the CDC
regarding all school-associated violent deaths similarly do
not show a steady increase or decrease.
8
Incidents in the current study took place in every month except July and
occurred on every day of the week except Sunday. The one incident that
occurred on a Saturday took place outside of a school prom.
Three-quarters of the attacks were carried out before the school day
began (n = 10, 24%) or during morning classroom hours (n = 21, 51%).
Firearm
61%
Knife
39%
WEAPON USED
*
*
Three attackers also used another
weapon including a hammer, a bo staff,
and a Molotov cocktail.
Timing of the Attacks
Before school .............................. 10 (24%)
Morning classroom hours .......... 20 (51%)
During lunch .....................................3 (7%)
Afternoon classroom hours ..........6 (15%)
Outside school hours ...................... 1 (2%)
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
10
BREAKS IN ATTENDANCE: Incidents occurred most
frequently at the start of the school year (Sep) or
after students returned from winter break (Jan).
Seventeen attacks (41%) took place within the
first week back to school following a break in
attendance, such as a suspension, school
holiday, or an absence due to illness or truancy.
Nearly one quarter of the attacks (n = 10, 24%)
took place on the first day that the attacker
returned to school after an absence. In two of
these incidents, the attacker was actively
suspended from the school at the time of the
attack. These ndings suggest that schools should
make concerted efforts to facilitate positive student engagement following discipline, including suspensions and
expulsions, and especially within the rst week that the student returns to school.
A 14-year-old student fatally stabbed a classmate at the middle school they both attended. At the time of the incident, the
attacker was suspended for truancy. Despite the suspension, witnesses reported seeing him walking the school’s hallways in
search of the target shortly before the stabbing took place.
Four attacks (10%) occurred on the rst day back from a regularly scheduled school holiday. Of those, three took place
on the rst day following an extended break, including the summer, fall, and winter holidays.
A 15-year-old student shot and wounded a random classmate in the high school
cafeteria on the rst day of the school year. The attacker had planned the attack over
the last week of his summer break. On the morning of the incident, he shared a message
on social media stating, “First day of school, last day of my life.” The shooting took place
four hours later.
DURATION OF THE ATTACKS: Most of the attacks (n = 34, 83%) lasted ve minutes
or less. Two-thirds of the incidents (n = 28, 68%) lasted for two minutes or less,
and nearly half of the attacks (n = 18, 44%) ended within one minute. Only six of
the attacks (15%) lasted longer than ve minutes, and none of the attacks lasted
longer than 15 minutes.
9
LOCATION OF THE ATTACKS: The attacks usually started and ended
in the same location (n = 36, 88%). The most common locations of
attacks were in classrooms and immediately outside of the school.
Other locations included cafeterias, hallways, and administrative
ofces. Attacks in restrooms, locker rooms, a gymnasium, and a
vestibule were less common.
Locations of the Attacks
Classroom Outside Office
Cafeteria Hallway Restroom
Locker room Gym Vestibule
DURATION OF THE ATTACKS
1 min
or less
44%
1-2 min
24%
2-5 min
15%
5-15 min
15%
Percentages may not total 100 due to rounding.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
Response Times
SCHOOL OFFICERS: In 27 cases (66%), a security ofcer or SRO was assigned to the school on either a full- or part-
time basis. During 20 of the attacks (49%), the ofcer was on duty at the school. In over one-quarter of the cases
(n = 12, 29%), the ofcer or SRO was able to make it to the scene of the attack within one minute. In three of the
attacks (7%), it took between one and ve minutes for the ofcer to respond, and for two attacks (5%), it took between
ve and ten minutes. Of note, in two cases the ofcers themselves were the targets of the attacks, and in the
remaining, it was not possible to determine the response time based on available information.
OUTSIDE LAW ENFORCEMENT: In just over one-third of the attacks (n = 16, 39%), outside law enforcement were
notified within one minute of the start of the attack, and in just under one-third (n = 12, 29%), outside police were
notied between one and ve minutes after the start of the attack. In nine cases, it took longer than ve minutes for
someone to notify outside law enforcement, and in four cases the timing could not be determined from
information available.
In nearly one-third of the cases (n = 13, 32%), it took external rst responders between one and ve minutes to arrive
on scene after the attack was initiated, and in about a quarter of the cases (n = 11, 27%) rst responders arrived
between ve and ten minutes after the attack began. In only one case, outside law enforcement responded in one
minute or less, because ofcers were already at the school conducting K-9 drug sweeps at the time of the attack.
Outside LE
Response
Time
Outside LE
Notification
School Officer
Response
Time*
1 (2%)
11 (27%)
3 (7%)
9 (22%)
16 (39%)
4 (10%)
––
4 (10%)
12 (29%)
2 (5%)
––
3 (7%)
1 min or less
>5 to 10 min
>15 min
Not found
13 (32%)
4 (10%)
––
12 (29%)
5 (12%)
––
3 (7%)
––
21 (51%)**
>1 to 5 min
>10 to 15 min
Not applicable
*Percentages may not equal 100 due to rounding.
**Includes those schools without ofcers assigned, assigned ofcers off-duty
at the time, and cases in which the SROs were targeted.
11
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
12
Resolution & Harm
ATTACK RESOLUTION: Half of the attackers (n = 21, 51%) ended the attack without any external intervention. Seven
attackers (17%) committed suicide, six (15%) left the scene, three (7%) surrendered to school ofcials, three (7%)
dropped their weapons and waited to be arrested, one (2%) stopped and called family, and one attacker (2%) left the
scene before calling family.
Non-law enforcement adult school staff brought nine attacks (22%) to an end. This included teachers, guidance
counselors, an assistant principal, a sports coach, a campus supervisor, and a janitor. Six of the attacks (15%) ended
with law enforcement intervention, either by SROs (n = 5, 12%) or by local police who were already on campus
(n = 1, 2%). Two of the attackers were killed by the law enforcement response. No attacks were ended by outside law
enforcement agencies responding to the scene from off-campus. Other attacks ended due to student bystander
intervention (n = 4, 10%) or a weapon malfunction (n = 1, 2%).
HARM: Ninety-eight victims were harmed in the 41 attacks, including 79 injured
and 19 killed. The victims included students, school staff, and law enforcement.
10
One stabbing incident accounted for 20 of the 98 victims, all of whom were injured
but survived. In just over half of the attacks (n = 22, 54%), only one individual was
harmed. In the remaining attacks, two persons were harmed (n = 9, 22%), or three
or more were harmed (n = 10, 24%).
TWENTY VICTIMS: A 16-year-old student randomly slashed and stabbed people at his high
school using two kitchen knives he had brought from home. Nineteen students and one
staff member were injured, but all survived. The attack ended after about ve minutes
when the assistant principal tackled the assailant.
Mass Attacks
The definition of a “mass attack,” as used by the U.S. Secret Service in its Mass Attacks in Public Spaces report series,
includes harm (i.e., injury or death) to three or more persons, not including the attacker. About one-quarter of attacks
(n = 10, 24%) in this study meet that definition of a mass attack.
PERSONS INJURED OR KILLED
Students
85%
Law Enforcement
2%
Staff
13%
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
Judicial Outcome
CRIMINAL CHARGES: Thirty-two attackers (78%) were criminally charged,
twenty-two of whom (54%) were charged as adults. Nearly all of those
charged as adults accepted plea agreements (17 pled guilty; 2 pled no
contest). Only two cases went to trial, where both attackers were found
guilty and sentenced as adults. A third attacker is still awaiting trial in
adult court.
Twelve attackers were processed through the juvenile justice system.
Of those charged as juveniles, the majority pled to their charges (9 pled
guilty; 1 pled no contest).
CURRENT DISPOSITIONS: Twenty-two of the attackers (54%) are
currently incarcerated and one attacker is a patient at a mental health
facility. Eight attackers (20%) have been released from juvenile or adult
correctional institutions.
11
Judicial Actions
Adult
Juvenile
No Charges Filed
Subject pled guilty
Subject pled guilty
Subject committed
suicide
Subject pled
no contest
Subject pled
no contest
Subject found guilty
Unknown
Subject fatally shot
Percentages exceed 100 as two attackers had both adult and
juvenile charges.
Charges led,
case pending
54%
29%
22%
22
12
9
41%
22%
17%
5%
2%
17
9
7
2
1
5%
5%
5%
2%
2
2
2
1
13
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
14
PART II: THE ATTACKERS
GENDER, RACE, & AGE: While the attackers were predominantly male (n = 34, 83%), seven of the attackers were
female (17%). Based on police report designations, approximately two-thirds of the attackers (n = 26, 63%) were
White, six were Black (15%), and two were Hispanic (5%). Their ages ranged from 12 to 18, with an average age of 15.
YOUNGEST: A 12-year-old student opened re at his middle school. He killed one teacher and injured
two classmates before fatally shooting himself. The attacker had a history of being bullied and felt
mistreated by students and teachers.
OLDEST: An 18-year-old former student, who had graduated the year prior, opened re outside of his
old high school toward a group of students exiting the prom. Two of the students were shot and
injured. Local police, already on scene conducting drug sweeps, instructed the attacker to drop his
weapon. When the attacker failed to comply with police instructions, he was fatally shot by one of the
ofcers. The attacker believed that his ex-girlfriend and her new boyfriend were attending the prom
together, but neither was actually in attendance
.
GRADE LEVEL: The grade levels of the attackers
ranged from 7th grade to 12th grade.
FORMER STUDENTS: Four of the attackers (10%)
targeted schools in which they were no longer
enrolled as students. They included a student who
attacked his former high school after transferring
to an alternative school; a student whose mother
withdrew him from school because he was about to be expelled; a student who had graduated the previous year, but
was working at the school as a teacher’s assistant and occasionally performed with the school marching band; and a
middle school student who attacked his former elementary school.
Male
83%
Female
17%
GENDER
Race/Ethnicity
White
American Indian or Alaska Native
Black/African American
Two or more races
Hispanic
Undetermined
63%
2%
26
1
15%
10%
6
4
Categories consistent with the U.S. Department of Education guidance on the
collection and reporting of racial and ethnic data. (72 Fed. Reg. 59267 Pub. 2007.)
5%
5%
2
2
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
15
Motives
When determining motives, all available information about each attacker's thinking and behavior in the time leading
up to the attack was considered, including explicit explanations of motive that the attackers provided through verbal
statements made before the attack, in suicide notes or manifestos, and during post-incident interviews with
law enforcement.
Consistent with other forms of targeted violence, the motives behind these 41 school attacks varied widely and were
based on the personal perceptions and experiences of the attackers. Most attackers had multiple motivations for
carrying out their attacks. This analysis indicates that a student’s motive to carry out a violent act at school is
usually multifaceted and is the byproduct of the student’s individual circumstances, as well as their personal
perception of those circumstances.
All but two of the 41 attackers had an identiable primary motive, dened as the motive that appeared to most
strongly contribute to the attacker’s decision to act violently. Most of the attackers (n = 35, 85%), however, had at
least one additional secondary motive that contributed to the decision for carrying out the attack.
SINGLE MOTIVE: A recently suspended 14-year-old student fatally stabbed a classmate in the chest as he exited his middle
school. The students had been friends, but their relationship had deteriorated due to frequent conicts. The victim of the
stabbing had previously threatened, harassed, and bullied the attacker. The attacker grew to fear for his life and stopped
attending school in order to avoid the victim, which led to the attacker’s suspension for missing class. The attacker later
reported that he had been motivated by fear, stating that if he had not returned to school to stab his classmate, he believed
that he would have been killed himself.
Components to Motive Primary Secondary Total
Grievances
Psychotic symptoms
Peers
Related to bullying
Staff
Romantic
Other personal
Unknown
Suicidal
Desire to kill
Total
Fame/Notoriety
25 (61%) 9 (22%) 34 (83%)
2 (5%) 3 (7%) 5 (12%)
15 (37%)
--
4 (10%)
4 (10%)
2 (5%)
11 (27%)
--
6 (15%)
5 (12%)
4 (10%)
26 (63%)
19 (46%)
10 (24%)
9 (22%)
6 (15%)
2 (5%) n/a 2 (5%)
3 (7%) 14 (34%) 17 (41%)
7(17%) 8 (20%) 15 (37%)
41 (100%)
2 (5%) 2 (5%) 4 (10%)
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
16
MULTIPLE MOTIVES: An 18-year-old student shot and killed a classmate he encountered immediately upon entering his school.
He then sought out the debate team coach, who escaped, before the attacker killed himself. The attacker appeared to have been
primarily motivated by a grievance with the debate team coach, who had removed him from a captain position on the debate
team. Additional motivations appeared to include grievances with other staff members and classmates, a sadistic desire to hurt
other people, and suicidal ideations.
GRIEVANCES: For most of the attackers (n = 34, 83%), retaliating for a grievance played a role in their motive. For
nearly two-thirds of the attackers (n = 25, 61%), it appeared to be their primary motive.
• Peers: Most frequently, grievances involved classmates (n = 26, 63%), and these peer grievances were usually
related to bullying in some way (n = 19, 46%). Other examples of peer grievances that did not involve bullying
included ongoing conicts and contentious relationships between students, and anger over a specic event or
situation among classmates.
A 14-year-old student fatally shot a classmate at his middle school. The victim had been the subject of harassment by the
attacker and other students, who would call the victim derogatory homophobic names. The attacker later reported that the
victim had made comments that made him uncomfortable, citing them as “the nal straw” in his decision to attack.
• Staff: In a quarter of the attacks, the attacker had a grievance that involved school staff (n = 10, 24%). For
four attackers (10%), this grievance with teachers or administrators was the primary motivation. In each of those
four incidents, the school staff members were specically targeted in the attack.
A 16-year-old student fatally stabbed his high school principal. According to his confession after the incident, the attacker
began planning his attack on the principal three months earlier, after he learned that he would be returning to the same high
school for 11th grade. He stated that the school principal made him the angriest, and he disliked the school and did not
want to attend anymore. He was also angry at having to follow school rules, for example, when the principal repeatedly
made him tuck in his shirt.
• Romantic: Nine cases (22%) involved a grievance related to a romantic relationship as a primary or secondary
motive. Two of these cases involved female attackers, while seven involved male attackers.
A 15-year-old student fatally shot a former romantic partner outside of their high school before committing suicide. The
students had dated for over two years, but the victim had recently informed the attacker that the relationship was ending.
According to media reports, the attacker had a history of suicidal ideations and depressive symptoms. In redacted
versions of the notes released publicly, the attacker expressed anger at the victim for wasting the attacker’s money, time,
affection, and “so much more.” The attacker also wrote, "We planned our future together," adding, "This week has been the
worst in my life," and, "All of this has destabilized me."
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
17
• Other Personal Issues: Six students (15%) were motivated by other personal grievances, and for two (5%), it was
their primary motive. These personal issues involved things like anger over a failed drug test at school and
retaliating for abuse or neglect occurring at home.
A 14-year-old student opened re toward random classmates in his high school cafeteria. Two students were shot
and injured, and two other students received minor injuries, possibly when struck by shrapnel. The attacker later
stated his motive was attributed to his troubled home life, including a distant relationship with his mother, conicts
with his father, and an addiction to Adderall.
DESIRE TO KILL: The second most prevalent motive among student attackers (n = 15, 37%) was a desire to kill,
evidenced by a sadistic interest in violence or violent topics, a desire for power over their victims, or the expectation
of joy or pleasure in causing physical harm to others. For seven attackers (17%), this appeared to be their primary
motive, and eight additional students (20%) experienced these types of thoughts as a secondary motive. Nearly all
attackers who were primarily motivated by a desire to kill targeted random victims (see Appendix A: Statistical
Analyses). While it is possible that a desire to kill could be associated with psychological symptoms, it often appeared
to develop as a response to the student’s personal circumstances and life experiences.
A 15-year-old student fatally shot a classmate at his high school. The attacker had been in previous conicts with the victim,
but he also displayed a fascination with causing physical harm to others. A few days before the incident, the attacker showed
a handgun and ammunition to classmates, highlighting that “these bullets bounce around inside of your body.” He was also
overheard saying that he was interested in seeing what it felt like to kill someone. During his subsequent trial, a clinical
psychologist said that the attacker had been “born to violent addicts, shoved into abusive foster homes, and largely ignored
by courts and caseworkers.”
SUICIDAL: Seventeen attackers (41%) were motivated by suicide to carry out their attacks. For three attackers (7%),
this appeared to be their primary motive; only one appeared to be motivated by suicide alone. An additional 14
attackers (34%) had suicidality as a secondary motive. These ndings reveal that suicidal ideations were rarely the
sole or primary factor in an attacker’s motivation for violence. Suicidal ideations were more typically found in
combination with, and secondary to, other motives.
A 16-year-old student intended to commit suicide-by-cop when he stabbed
his SRO seven times with a bayonet. The ofcer struggled with the
attacker before fatally shooting him in self-defense. The attacker’s nal
words to the ofcer were, “Thank you sir, thank you.” In a suicide note, the
attacker described human civilization as an “unnatural system,” and
included additional language like, “If you’re reading this, I’m either dead or
on my way to dying. If I fail at suicide, something else will end me. I have
had enough of this tainted and corrupted world we live in,” and, “I hope for
a better existence after my passing.”
Youth Suicide
In 2017, the CDC reported suicide as the second-
leading cause of death among those between the
ages of 10-24 years old.
12
The National Suicide Prevention Lifeline provides
free and condential 24/7 support to people in
crisis. If you or someone you know needs help, call
1-800-273-TALK (8255).
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
18
FAME/NOTORIETY: Four students (10%) were motivated by a desire for fame or notoriety as a primary or secondary
motive. Each of these four students expressed a desire to emulate previous mass attackers, including those who
perpetrated the attacks at Columbine High School, Virginia Tech, and Sandy Hook Elementary School.
A 14-year-old student opened re at his SRO in an effort to steal the ofcer’s weapon. He intended to use the ofcer’s rearm
to perpetrate a mass shooting at his high school. The ofcer was burned by the discharging rearm, but detained the student
before further harm occurred. Three months prior, the attacker began keeping an extensive journal about his plans, in which
he described his research into the attacks at Columbine High School and Virginia Tech. He wrote about his desire to be
“worshipped” like previous attackers, adding, “I hope to out-do my idols and leave my mark on the world.” He wrote about how
he hoped that security camera footage of his attack would be on YouTube for his “followers” to study. After the attack, while in
juvenile detention, the student continued to express interest in the media coverage of his attack, asking staff at the detention
center if he could read the comments posted by readers of the online news articles.
PSYCHOTIC SYMPTOMS: Five students (12%) appeared to be motivated by psychotic symptoms, either as a primary
or secondary motive. These cases involved such symptoms as auditory hallucinations instructing the student to hurt
people, compulsive homicidal thoughts and visions, and a delusional belief that the student was in a romantic
relationship with a classmate.
A 15-year-old student randomly stabbed two classmates in a high school restroom. Both victims survived the attack. The
attacker had a history of homicidal and suicidal thoughts, thoughts of violence toward her family and girls at school, and had
previously threatened to stab a friend’s boyfriend. According to court records, six months prior to the attack, she was admitted
to an outpatient hospitalization program for her mental health. One counselor asked the attacker if she had suicidal or
homicidal ideations, and she responded, “Sometimes I get really sick thoughts. Sick ways of killing them, stabbing them,
slitting their throats, sexually abusing people at school.”
Targeting
In all forms of targeted violence, the targeted person or persons are usually selected by the attacker in relation to his
or her motive. For example, if an attacker has a grievance with a specic person or group, they may target that person
or group during the attack. In many of those cases, however, additional bystanders may be injured or killed, either as
unintended collateral victims or as victims of opportunity. In other instances, the attacker may indiscriminately target a
population of people, like the entire student body at a school, depending on the attacker’s goals and motives.
SPECIFIC PERSON TARGETED: In nearly three-quarters of the attacks (n = 30, 73%), the attacker targeted a specic
person or persons. In 22 attacks (54%), a specically targeted person was injured or killed.
A 12-year-old student stabbed her middle school classmate in the neck and stomach, injuring him. Other students had
previously observed the victim teasing the attacker and calling her names. A few days before the incident, the attacker told
another student that she hated the target. Just minutes before the stabbing, the attacker told a group of classmates that she
hated the target and that she wanted to kill him because he was annoying and was a bad person. Nobody else was targeted
or harmed in the attack.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
SPECIFIC GROUP TARGETED: In six attacks (15%), a specic group of students was targeted (e.g., bullies, “jocks,”
“emos,” “preps,” “sinners”). In only one of those attacks was a member of the targeted group actually harmed.
RANDOM VICTIMS: Seventeen attacks (41%) involved random victims who were attacked indiscriminately or
opportunistically by the attacker (e.g., the attacker red into a crowd or red upon the rst student he saw).
A 16-year-old student randomly attacked a female classmate with a serrated knife and a claw hammer when the victim
walked into the high school restroom. The victim was stabbed and hit several times but survived the attack. The attacker’s goal
was to kill as many people as she could before eeing the school. She then planned to purchase a gun and commit suicide.
COLLATERAL VICTIMS: Five attacks (12%) involved random victims who were injured as collateral bystanders (e.g., the
attacker aimed at a specic target, but an unintended student was shot).
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
19
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
20
Planning
Over half of the attackers (n = 21, 51%) engaged in observable planning behaviors prior to carrying out their attack
that went beyond making statements of intent. Twelve attackers (29%) exhibited three or more different types of
planning activities that were observable to others. Of note, the attackers in this study who perpetrated mass attacks
averaged more than twice as many planning behaviors as attackers with fewer victims (see Appendix A: Statistical
Analyses). For 10 of the attackers (24%), the rst act of planning occurred within one month of the incident. Seven
attackers (17%) began planning between one and six months before the attack, and rarely did planning begin more
than six months prior (n = 3, 7%).
13
While some attackers discussed their plans with others, no attackers appeared to
receive direct assistance in planning their specic school attack.
WEAPONS RESEARCH AND SELECTION: One-third of the attackers
(n = 13, 32%) researched weapons prior to the attack. This
included conducting online searches, reading books and
pamphlets, and asking others about weapons experience and
other related topics. The research focused on how to make
explosives, weapon functionality and utility, how to inict maximum
damage, and types of protective equipment.
DECEPTIVE PRACTICES: Nine attackers (22%) engaged in deceptive
practices in order to hide their activities and avoid detection. This
included lying about weapons and other materials (e.g., telling
parents that purchased materials were for chemistry class),
concealing weapons for transport, claiming needing a weapon for
protection, and waiting until after dark to manufacture weapons. For example, on the night prior to one attack, the
student waited until his mother went to sleep before building homemade explosives. He then pretended to be asleep
so that his mother could wake him up to start the day, thus maintaining their normal routine and not arousing
any suspicion.
WEAPONS-RELATED: Seven attackers (17%) engaged in weapons-related planning behaviors, in addition to weapons
research (see Weapons Research and Selection) and the acquisition of the weapons used in the attack (see Firearm
Acquisition section). These other activities included manufacturing explosives, modifying guns or knives, stockpiling
additional weapons, and failed attempts to acquire other weapons. Some who failed to acquire certain weapons or
materials adjusted their plans to use other weapons already acquired or more readily accessible.
A 12-year-old student shot and wounded several classmates and a school security ofcer in his middle school gymnasium
before surrendering to a teacher. Two days prior to the incident, the attacker had attempted to acquire his father’s handgun
but was unsuccessful. The night before the attack, he took a shotgun from his home and sawed off the barrel so that it
could t into his duffel bag.
Planning for the Attack
Weapons research Planning the
and selection execution of the
attack
Deceptive practices
Practicing with a
Weapons-related weapon
Documentation of Recruiting others
plans
Packing an attack
Approaching, bag
surveilling, or
researching the Researching prior
target attacks
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
21
DOCUMENTATION OF PLANS: Seven attackers (17%) documented their plans and intentions for the attack in journal
entries or other writings. These writings included hit lists, safe lists (i.e., lists of those individuals who would not be
targeted), supply lists, lessons learned from past attackers, and diagrams of school hallways.
A 15-year-old student fatally shot a classmate and injured three others at his high school before surrendering to the school
janitor. In his journal and other scattered writings, police found diagrams mapping his high school, along with written
homicidal statements and lists of chemicals, weapons, and ammunition. The attacker had also written “X Kill” on the photo
page in his yearbook and had crossed out the photographs of 26 classmates.
APPROACHING, SURVEILLING, OR RESEARCHING THE TARGET: Five of the attackers (12%) researched their targets
prior to the attack. This included surveilling the SRO in order to learn his routine, noting security camera locations,
and trying to arrange meetings with a targeted teacher.
A 14-year-old student shot and killed his father before driving to his former elementary school, where he opened re near the
playground. One elementary student was fatally shot, and at least one other student and one teacher were injured. The
attacker had researched the schedules of schools he had attended and inquired about security and the presence of SROs. He
chose to target his former elementary school due to the more robust security at his middle school, his understanding of police
response times to different schools, and the proximity of the elementary school to his home.
PLANNING THE EXECUTION OF THE ATTACK: Five of the attackers (12%) spent time carefully considering how they
would execute their attacks. They considered the “best” location to kill their intended target(s), the class period to
initiate the attack, and the route they would take through the school. One of the attackers shared his plan with
friends via social media.
RESEARCHING PRIOR ATTACKS: Five of the attackers (12%) conducted research on prior incidents of targeted
violence, including past school attacks. This research was usually conducted through online searches, or by
watching videos or documentaries of prior attacks.
A 16-year-old student used two kitchen knives to injure 19 classmates and a school security guard. The attacker selected the
birthday of one of the Columbine attackers as the date of his attack. Prior to the incident, the attacker had watched a
documentary about the Columbine shooting in an effort to “numb himself” in preparation for his planned attack.
PRACTICING WITH A WEAPON: Four attackers (10%) practiced with a weapon prior to the attack. This included one
attacker who tested his homemade pipe bombs, one who practiced loading and unloading a shotgun, and one who
practiced wearing tactical gear.
RECRUITING OTHERS: Four attackers (10%) contemplated or attempted recruiting others to help execute the acts of
violence. Two of these attackers attempted unsuccessfully to recruit a specic individual to assist with the attack.
Two others considered recruiting help but did not. One of those attackers realized that attempts for recruitment
could make him vulnerable to detection and therefore refrained from asking.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
22
PACKING AN ATTACK BAG: Two attackers (5%) packed a supply bag prior to the incident. One of them packed
clothes, cash, and other items needed to run away from home after the attack, and the other packed additional
weapons, an energy drink, and a lollipop.
OTHER TYPES OF PLANNING: Nine attackers (22%) took steps to plan the attack in ways that were unique and
could not be categorized among the other behaviors. This included stealing money to buy weapons, deleting an
email account, and specically wearing baggy pants the day of the attack in order to conceal a weapon.
Firearm Acquisition
The primary weapons used in the 41 attacks were rearms (n = 25, 61%) and knives (n = 16, 39%). In most of the
incidents involving knives, the weapons were taken by the attackers from their kitchens or other areas of their homes.
Because knives are commonly available with little restriction, this section focuses on how the attackers acquired the
firearms used in the attacks. All percentages provided in this section are based on the 25 attackers who
used firearms.
FROM THE HOME: Nineteen attackers (76%) acquired a rearm from the home of a parent or another close relative. In
half of the firearms cases (n = 12, 48%), evidence indicates the firearm was either readily accessible, or it was not
secured in a meaningful way. For example, some rearms had been kept locked in accessible wooden or glass
cabinets, locked in vehicles, or hidden in closets. In four cases (16%), the firearms were kept in more secured
locations, but the attacker was still able to gain access to them. In these instances, the rearms were secured in
a locked gun safe or case, but the attackers were able to gain access to them because they knew the combination or
where the keys were kept, or they were able to guess the password or
combination. In three cases, it is unknown if the rearm had been secured.
A 15-year-old student fatally shot one classmate and injured three others at
his high school. The attacker knew the combination to his father’s gun safe,
from which he was able to obtain a .32 caliber semiautomatic pistol and a
semiautomatic .223 caliber AR-15 rie with a 30-round magazine. The attacker
transported the weapons on the school bus by concealing the rie and seven
extra boxes of ammunition in a golf bag. He kept the handgun in his pocket.
TIMING OF THE ACQUISITION: Eight of the attackers (32%) acquired a rearm
on the day of the attack. Five additional attackers (20%) acquired a rearm
the day before the attack, and four attackers (16%) acquired a rearm
between two and seven days prior. This nding reinforces the importance
of a swift response to situations involving students who may pose a risk of
harm to themselves or others, especially those who have access to
weapons in the home.
Juvenile Access to Weapons
Under federal law, individuals under the age
of 18 may not legally possess a handgun,
except in limited circumstances.
Federal law does not restrict the age of
individuals who may possess long guns
(e.g., shotguns and ries). Some states
have implemented additional restrictions on
juvenile possession of rearms, for example,
prohibiting minors from possessing any
rearm, except for certain activities and with
parental consent.
14
If a child poses a risk of harm to him/herself
or others, it is the responsibility of parents,
law enforcement, and schools to
collaboratively determine the most
appropriate avenue for ensuring that the
child does not have access to weapons.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
23
WEAPONS TRANSPORT: Half of the attackers (n = 12, 48%) who used a rearm in their attack transported the
weapon into the school building while concealing it in their backpacks. Three attackers (12%) transported their
rearms into the school in another type of bag, including a guitar case, a golf bag, and a gym bag. One attacker
transported a rearm into the school concealed in his waistband. Two attackers did not conceal their rearms,
carrying them openly into the school building after the start of the school day.
Additionally, two attackers transported rearms from home to school but perpetrated the attack on school property
without entering the building. One of those attackers carried his weapon out in the open, while the other concealed
the rearm in a guitar case as he rode his bike to the school grounds.
Law Enforcement Contact
Half of the attackers (n = 17, 49%) had contact with law enforcement prior to carrying out their attacks, which included
prior arrests and other encounters, like welfare checks at home. For some (n = 6, 17%), these prior contacts or
interactions were with an SRO.
PAST CRIMINAL CHARGES: About one-third of attackers (n = 11, 31%) had previously been arrested or faced criminal
charges, and the crimes for which they were charged included both nonviolent (n = 7, 20%) and violent (n = 6, 17%)
offenses. In some of these cases, the charges against the attacker were ultimately dismissed. The nonviolent charges
included alcohol- or drug-related offenses, curfew violation, criminal conspiracy, shoplifting or theft, trespassing, and
reckless driving. The violent charges included affray, assault with a deadly weapon, and weapons possession at school
with intent to cause harm. Physical ghts, which occurred both at home and at school, resulted in charges of
disorderly conduct and disrupting public school. In one case, the charge was related to a domestic violence incident
when the attacker physically assaulted his mother, threatened to kill her, and ended up having to undergo a mental
health evaluation. None of the charges were for sex offenses.
An 18-year-old student shot and injured a classmate at his high school before surrendering to the SRO. According to
investigative records, the attacker had been previously arrested for carrying a concealed gun, possession with the intent to
distribute a controlled dangerous substance, trespassing, and resisting a public ofcer.
OTHER CONTACT WITH LAW ENFORCEMENT: Some attackers (n = 10, 29%) had prior interactions with law enforcement
that did not result in arrests or charges or were related to issues with their families. These types of interactions
included instances where the attacker had gotten into ghts or displayed other inappropriate behavior at school and
was addressed by the SRO as a result. In one case, the prior contact was part of a threat assessment conducted by
For 6 of the 41 incidents included in this study, there was little information available on the backgrounds of
the attackers in both primary source les and open sources. As a result, those six attackers were removed
from the analysis of behavioral histories. All percentages provided in the remainder of the report are based
on the 35 attackers for whom substantive information was available.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
24
school ofcials during which the SRO searched the attacker’s home for weapons. Prior contact with law enforcement
also involved an attacker being detained by law enforcement after being reported as a runaway and police visits to
attackers’ homes for issues regarding the attackers’ families.
Psychological, Behavioral, & Developmental Health
According to national prevalence rates, nearly 20% of children are diagnosed with at least one mental health and/or
behavioral disorder.
15
In this study, a documented mental health diagnosis was received by fourteen (40%) of the
attackers prior to their attacks.
This section will focus on the observable mental health symptoms displayed by attackers prior to their attacks. Such
mental health factors were divided into three main categories: psychological, behavioral, and neurological/
developmental. Most of the attackers (n = 32, 91%) in this study exhibited symptoms in at least one of these
categories, and half of the attackers (n = 17, 49%) displayed symptoms of more than one type, with the most frequent
combinations involving psychological and behavioral symptoms. When considering such signs and symptoms, it is
possible that underlying situational factors (e.g., stressors the child is experiencing) may be the cause of the
behavior, as opposed to a diagnosable disorder.
Psychological
This section describes attackers who exhibited several clusters of
emotional and psychological symptoms. The most common of these
types of symptoms was related to depression, with two-thirds of the
attackers (n = 22, 63%) exhibiting some sign or symptom. A similar
number of attackers (n = 21, 60%) experienced suicidal thoughts.
16
Anxiety (n = 10, 29%), anger (n = 9, 26%), and symptoms of
psychosis (n = 7, 20%) were also documented in the available
sources of information.
Symptom Categories
Psychological
Neurological/Developmental
Behavioral
24 69%
7 20%
20 57%
The vast majority of individuals in the United States displaying the symptoms or traits identied in this section
do not commit acts of crime or violence. The symptoms described in this section constitute potential
contributing factors, as part of a constellation of other factors, and should not be viewed as causal explanations
for why the attacks occurred.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
25
A 16-year-old student shot his high school principal in the arm before his weapon malfunctioned. Prior to the incident, the
attacker exhibited a wide range of symptoms typically associated with depression. For example, he began to isolate himself
after his parents divorced and his siblings moved out of the home. While he had played multiple sports and enjoyed video
games, the attacker withdrew from these activities as his depression worsened. He began going to school without shaving,
showering, or washing his hair, and multiple people noted his body odor. The student reported feeling helpless, hopeless, and
worthless, and he said he had low energy, trouble sleeping, reduced appetite, decreased interest in activities, and impaired
concentration. His grades began to decline. He often sat alone in his room in the dark. He began having suicidal thoughts
and came close to killing himself prior to the incident. According to media reports, a state psychiatrist concluded that mental
illness played a role in the attacker's actions in carrying out the incident.
Behavioral
Twenty attackers (57%) displayed behavioral symptoms,
including deance toward authority, poor impulse control,
and other violations of social norms. Also included here
are those attackers who exhibited symptoms of, or were
diagnosed with, ADD or ADHD. Anger was included here if it
occurred in the absence of any identiable psychological or
emotional issues that would account for it.
A 15-year-old student shot and injured two classmates at his high school. The attacker had a history of misconduct beginning
at a young age. His academic record included dozens of discipline referrals at school for disobedience, disrespect to teachers,
threats, harassment, and assault. He had numerous in-school and out-of-school suspensions. In sixth grade alone, he had 24
discipline referrals from ve different staff members. The reasons for the referrals included refusing to cooperate with
authority and altercations with peers. He was viewed as a bully and repeatedly tried to pick a ght with the boy he later
targeted in the shooting
.
Neurological/Developmental
Neurological conditions or developmental issues were found for seven attackers (20%), including developmental
delays, cognitive decits (e.g., poor problem-solving or decision-making skills), learning disabilities (e.g., dyslexia), and
symptoms or diagnoses within the autism spectrum.
Attackers in this category often experienced delays in language
acquisition (e.g., lifelong speech challenges, not conversing
until ve years old), sensory sensitivity (e.g., covering ears
when other kids were loud, disliking school assemblies, or
attempting to avoid pep rallies due to noise), and poor
communication skills (e.g., saying things that did not make
sense). In addition, these students often misinterpreted social
cues or isolated themselves. With an inability to integrate
Symptom/Behavior
Deance/Misconduct
Anger
Aggression
ADD/ADHD
Animal cruelty
14
5
40%
14%
8 23%
10
3
29%
9%
Neurological/Developmental Symptoms
Sensory sensitivity Cognitive deficits
Poor communication Learning disabilities
Developmental delays Poor social skills
Delayed language acquisition
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
26
socially and communicate effectively, most of the attackers who experienced these conditions had difculty regulating
their emotions while also experiencing other psychological and/or behavioral symptoms.
A 12-year-old student fatally shot one teacher and wounded two classmates at his middle school before killing himself.
According to open source reporting, the attacker was late in his language development and required speech therapy. As he
progressed through elementary school he continued to have problems with communication, cognitive decits, and academics.
In sixth grade, he was still performing rst-grade work in math and second-grade work in reading. The attacker exhibited poor
social skills and did not make friends easily. His ability to pick up on social cues was limited, and he had difculty with
transitions from one activity to another. His emotions were unstable, with frequent episodes of anxiety, anger, and crying. He
often seemed confused and was described as being in his own world
.
Personality Traits
A fourth category, related to personality traits, was also considered in order to highlight the personality dynamics that
may have impacted the attackers’ behavior. A personality disorder is evidenced by a manner of thinking, feeling, or
behaving that falls outside of social norms and causes distress or problems functioning. It is possible that persistent
patterns of deance and misconduct at a young age may be indicative of a personality disorder as a student reaches
adulthood. Research shows that as many as two-thirds of adolescents with a diagnosis of conduct disorder can go on
to develop a personality disorder as an adult.
17
One notable feature of attackers in this study with possible personality traits was narcissism, which manifested in
multiple ways. These attackers generally had an inated sense of self, believing that they were superior to others.
Because they viewed themselves as superior, they had little concern for rules or laws, believing that they were above
such conventions. Similarly, they did not respect people in positions of authority over them. Some of them sought
fame through violence as a way to validate their inated self-concepts. Furthermore, these attackers tended to lack
empathy and derived sadistic pleasure from the thought of hurting or killing others. Further evidence of callousness
was seen in their lack of remorse or emotional reaction to the enormity of what they had done.
A student shot and injured a classmate and a teacher in his
high school science classroom. The attacker had previously
written stories and told friends about his desire to attack the
school and his bullies. Prior to the incident, the attacker had
expressed fantasies of killing and mutilating his brother, pushing
a child out of a wheelchair, and punching a crying infant.
Classmates also observed the attacker drawing portrayals of
children being massacred on a playground. After the incident,
investigators found journals with similar violent content. During trial,
a psychologist retained by the prosecution stated that the attacker
had antisocial, paranoid, and narcissistic features.
Personality Traits
Narcissistic Sadistic
Deceitful Disregard for rules
Lacks empathy Callous
Manipulative
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
27
Mental Health Treatment
Of the 35 attackers in this study, 19 (54%) had received some type of mental health treatment. Eleven (31%) had
received services within their schools, and half (n = 17, 49%) had received services in the community. Nine of the
attackers (26%) received services both at school and in the community. For some attackers, mental health treatment
was in place at the time of the attack, while for others, the treatment had ended by the time the incident occurred.
The fact that half of the attackers had received one or more mental health services prior to their attack indicates that
mental health evaluations and treatments should be considered a component of a multidisciplinary threat
assessment, but not a replacement. Keep in mind that standard mental health evaluations and treatments will not
include things like a review of criminal records, corroborative interviews with people who know the student, an analysis
of a student’s social media accounts, or a search of the student’s home or locker. Mental health professionals, both
in and out of the school, should be included in a collaborative threat assessment process that also involves teachers,
administrators, and law enforcement.
Substance Use & Abuse
Half of the attackers (n = 17, 49%) had a history of substance use and/or abuse. The most frequently used substances
included marijuana (n = 14, 40%) and alcohol (n = 12, 34%). Some attackers used other controlled substances
(n = 7, 20%), including opiates (e.g., Percocet, oxycodone), amphetamines (e.g., Adderall), Valium, Seroquel, K2, and
anabolic steroids, none of which had been prescribed to them. A few attackers (n = 3, 9%) misused drugs that had been
prescribed by a doctor, or over-the-counter drugs, including Benadryl and cough syrup with codeine.
For seven of the attackers (20%), their substance use reached the level of abuse, causing negative consequences or
problems in their lives such as alcohol- or drug-related criminal charges, school suspensions, or being sent to drug
treatment.
A 16-year-old student stabbed and injured a classmate at his high school. Other classmates subdued the attacker until school
ofcials arrived at the scene. According to court records, around the age of 13 or 14, the attacker began using marijuana and
alcohol with his friends, as well as using non-prescribed oral steroids and abusing cough syrup containing codeine. The
attacker reported that he consumed a four-ounce bottle of the cough syrup every two days for about a year.
Weapons Use & Violence
Weapons Use: Three-quarters of attackers (n = 27, 77%) had a history of using weapons prior to the attack. These
histories included the attackers owning, using, training with, or practicing with weapons, including rearms, knives, or
other types. For example, some attackers were experienced in hunting or target shooting, were members of the Junior
Reserve Ofcer Training Corps (JROTC), had interests in airsoft guns, regularly carried a knife, or played with BB guns.
While each of these activities, independently, do not indicate that a student poses a risk of violence, they do provide
the student with familiarity, and in some cases explicit training, on how to handle and use weapons.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
28
Twenty-ve attackers (71%) had previous experience with rearms,
including pistols, ries, shotguns, and other weapons that closely
mimicked rearms, such as BB guns, airsoft guns, and paintball
guns. Thirteen attackers (37%) had previous experience with knives,
outside of normal usage, and eleven (31%) had previous experience
with both rearms and knives. Eight attackers (23%) had
experience with other types of weapons, which included
manufacturing explosives (e.g., pipe bombs) or incendiary devices
(e.g., Molotov cocktails), and using a bow and arrow, nunchucks,
throwing stars, or a machete.
A 15-year-old student fatally shot one classmate and wounded one teacher before killing himself. In his language arts
notebook, the attacker had made several entries about weapons. He wrote, “Weapons-I know A LOT! About guns, knives, and
body armor. I can tell the name of a gun… and tell you it’s [sic] caliber. I can also disassemble two guns.” The attacker’s
brother had taken him shooting in the year prior to the attack, where they practiced shooting pellet guns, an AR-15 style rie,
and a rie the attacker had received as a gift from his brother for his birthday. During that same year, the attacker
occasionally brought knives to school and showed them to at least one other student on the school bus.
HISTORY OF VIOLENCE: Half of the attackers (n = 18, 51%) engaged in violent behaviors that did not result in an
arrest or contact with law enforcement, including violence toward a family member (e.g., hitting a parent or stabbing a
sibling), physical assaults (e.g., choking a friend or hitting another student), threatening or aggressive behavior (e.g.,
pointing a pair of scissors at and threatening to stab a classmate), and hurting animals. Additional cases of attackers
with a history of violent arrests or criminal charges are described in the Law Enforcement Contact section of
this report.
A 14-year-old student killed his father at home before opening re on his former elementary school. One elementary student
was fatally shot, and at least one other student and one teacher were injured. The attacker had a history of hurting small
animals and watching videos of animals being hurt.
INTEREST IN VIOLENCE AND WEAPONS: Half of the attackers (n = 17, 49%) had an unusual or concerning interest in
violence or weapons. For example, some of the attackers researched past incidents of mass violence, drew gures of
dead students, consumed violent or graphic media, completed writing assignments or kept journals referencing violent
topics, hurt animals or watched videos of animals being abused, or described themselves as obsessed with weapons.
For an interest in weapons to be concerning, it had to be viewed as a xation or fascination that was beyond what
would be considered a normal interest or knowledge of weapons.
A 12-year-old student fatally shot a teacher and wounded two classmates at his middle school before killing himself. In the
months preceding his attack, the attacker conducted Internet searches using the following terms: bullying, Top 10 evil children,
Super Columbine Massacre Role Playing Game, shoot, guns, revenge, murder, school shooting, killer, hate, what if Nazis
won WW2. The attacker also owned 47 shooter video games and had saved photographs of the Columbine shooters on
his phone.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
29
Two specic interests emerged as common and particularly concerning among the attackers. These were a fascination
with the Columbine attack and an interest in Hitler and Nazism.
• Columbine: About one-quarter (n = 8, 23%) of attackers displayed a concerning interest in the attack at Columbine
High School. This was demonstrated by the attacker having saved photos of the Columbine attackers, watched
documentaries or read about the attack, or paraphrased the diary of one of the Columbine attackers in their own
diary. These attackers also consumed the movies or music that they knew the Columbine attackers had liked, spoke
to friends about the attack, posted about it on social media, and wrote about Columbine in class assignments,
journals, or manifestos.
A 14-year-old student opened re toward his SRO, burning the ofcer on the head with the discharging rearm. The attacker
had previously written a report for class titled, “The Columbine High School Massacre.” Throughout his personal journal, he
referenced the Columbine attackers, referring to them by nicknames, and also referenced a movie they liked. He wrote,
“[The Columbine] story is SO fascinating to me!” and, “They’re so... INSPIRATIONAL!”
• Hitler/Nazism: Seven attackers (20%) displayed a concerning interest in Hitler or Nazism. This interest was revealed
through a pattern of behavior, including attackers who turned in school assignments describing their admiration
for Hitler, collected Nazi propaganda and Hitler speeches, used known Nazi references and symbols, viewed
neo-Nazi websites, wrote about Hitler or Nazism in journals, or had conversations with classmates about
these topics.
A 14-year-old student fatally shot a classmate at his high school. The attacker was known to have shown an interest in Hitler,
and he knew the names of all of Hitler’s deputies. After the attack, a copy of Hitler's Mein Kampf was discovered in the
attacker’s backpack along with drawings of swastikas in his notebooks. A later search of the attacker’s residence identied
Nazi paraphernalia, Nazi and white supremacist materials, anti-Semitic drawings, Nazi videos, Hitler speeches, and Nazi
propaganda posters from Internet websites.
Home Life Factors
This section is intended to describe the family dynamics and home experiences for the 35 attackers, using several
home life factors that emerged as common and impactful. While none of the factors included here should be viewed
as predictors that a student will be violent, the eld of study examining adverse childhood experiences (ACEs) nds
that experiencing certain adverse events in childhood can result in a range of negative outcomes later in life,
particularly when a child experiences multiple ACEs.
18
ACEs include things like parent separation and divorce,
substance abuse in the home, family violence, abuse, family incarceration, and family mental illness. The rst analysis
of nationally-representative data on ACEs found that the percentage of youth displaying indicators of poor wellbeing is
greater among those who have experienced three or more such factors.
19
In other words, the more ACEs a child
experiences, the greater the risk for negative outcomes for the child.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
30
All but two of the attackers (n = 33, 94%) experienced at least one of the home life factors listed here. Those
attackers experienced between one and six factors each, with an average of 3.4.
SEPARATION/DIVORCE: The most common home life factor experienced was parent separation or divorce, which was
experienced by 25 of the attackers (71%).
FINANCIAL DIFFICULTY: Twenty-four attackers (69%) experienced a home life with at
least one indicator of nancial difculty.
ARRESTS OR INCARCERATIONS: Nineteen attackers (54%) had a home life where a
parent, guardian, or sibling was arrested or incarcerated at some point after the
attacker’s birth. Research has found that the arrest or incarceration of a family
member can have a number of negative consequences for childhood development.
Children who witness a parent’s criminal activity, arrest, and incarceration have a
greater likelihood of emotional maladjustment and a harder time regulating
their emotions.
20
A 15-year-old student fatally shot a classmate at her high school. She then ed the
scene before turning herself in. At the time of the incident, the attacker’s father was
serving a 25-year sentence for second-degree murder. He was arrested two years
prior to the attack. While the student had not visited her father in prison, she was
able to speak with him on the phone and had lived with him intermittently prior to
his incarceration.
SUBSTANCE ABUSE: Sixteen attackers (46%) had parents, guardians, or siblings using or abusing illicit drugs or
alcohol. Research has found that children of parents with substance use disorders are more likely to experience
abuse or neglect, develop behavioral problems, display violent behaviors, or engage in substance use themselves.
21
Financial Difculty
Bankruptcy
Eviction or homelessness
Failure to pay
child support
Foreclosure
Fraudulent check(s)
Lien
Low income
Poverty
Home Life Factors
Parents separated/divorced
Family discord (including domestic violence/abuse)
Family mental health
Parent(s) or sibling(s) arrested/incarcerated
Non-parental custody/care
Family nancial difculty
Abuse/neglect suffered by the attacker
Parent(s) or sibling(s) substance abuse
25 71%
14 40%
8 23%
19
4
54%
11%
24
8
69%
23%
16 46%
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
31
A 15-year-old student stabbed two classmates multiple times in the restroom. Both victims, selected at random, survived
the attack. When she was a child, the attacker’s father was incarcerated multiple times for drug possession and other
drug-related offenses. On at least one occasion prior to the attack, according to open source reporting, the attacker snuck out
of her mother’s home in order to smoke marijuana with her father.
FAMILY DISCORD: Fourteen attackers (40%) had evidence of exposure to family discord. The types of discord captured
here varied, from repeated verbal arguments to domestic and other physical violence. These hostile events often
occurred multiple times throughout the attackers’ lives, and most of these attackers experienced more than one
type of family discord.
A 17-year-old student entered his former high school and fatally shot three classmates and wounded three others. The
attacker came from a troubled home with both of his parents involved in multiple unions and divorces. According to public
records, both parents also had a history of criminal charges and substance abuse. Domestic violence was common in the
home, and the parents’ criminal behavior resulted in the attacker ultimately being raised by his grandparents.
FAMILY MENTAL HEALTH: Eight attackers (23%) had evidence of living with a family member in the home who
experienced symptoms of a mental illness. The most prevalent conditions appeared to involve mood disorders
(e.g., depression and bipolar disorder), with a few instances of possible psychosis.
ABUSE/NEGLECT: Eight attackers (23%) suffered from neglect or were the victims of physical, verbal, or sexual abuse
in the home. Physical abuse was the most common type of abuse, and ve attackers experienced multiple types of
abuse prior to their attacks.
NON-PARENTAL CUSTODY/CARE: Four attackers (11%) were raised by someone other than a biological parent at some
point in their lives. Three had lived with, or were raised by, their grandparents. One attacker had been placed in the
foster care system and spent time in multiple foster homes before he was adopted.
Stressors
It is widely recognized that high stress levels are associated with emotional and behavioral problems for children.
22
All 35 attackers (100%) experienced stressors, with nearly all (n = 33, 94%) experiencing at least one in the six
months prior to their attack. Three-quarters of the attackers (n = 26, 74%) experienced a stressor within one month
of the attack, and half of the attackers (n = 18, 51%) did so within two days. Research indicates that some types of
stress, like the loss of a loved one, can be tolerated by children if they have emotional supports to help them process
the experience.
23
Even with those supports, however, children will be challenged if the stressors they experience are
intense or frequent enough. In such cases, children may exhibit a range of responses, from depression
to hyperactivity.
24
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
32
SOCIAL: All of the 35 attackers (100%) in this study experienced at least one social stressor. Social stressors
identied in this analysis included stress related to the attackers’ relationships with peers (e.g., bullying or other peer
conicts) and romantic partners. These types of stressors in particular, including peer rejection during childhood and
adolescence, have previously been linked to depression, anxiety, substance use, and behavior problems.
25
For half of
the attackers (n = 19, 54%), a social event, such as a breakup or an incident of bullying, was the most recent stressor
experienced before carrying out the attack. Those stressors often occurred within a week prior (n = 11, 31%).
A 16-year-old student stabbed and injured his classmate in the high school cafeteria with a knife. The attacker later described
himself as having experienced chronic verbal and physical bullying throughout his childhood and adolescence. In the weeks
leading up to the attack, the victim of the stabbing began to date the attacker’s ex-girlfriend, which angered the attacker.
FAMILY: Stressors related to the attackers’ families included conicts in the home, abuse or neglect by a parent, and
family nancial difculties. Nearly all attackers experienced family-related stressors (n = 32, 91%). At least three
attackers had a stressful interaction with a parent within one day of the attack.
A 14-year-old student fatally shot one elementary school student and injured at least one other student and one teacher
outside of his former elementary school. According to the attacker, his father had been angry and aggressive over the previous
24 hours, complaining about his nances and engaging in a confrontation with the attacker over homework. The attacker
fatally shot his father before the attack at the school.
ACADEMIC/DISCIPLINARY: Most attackers (n = 31, 89%) had experienced school stressors related to academic or
disciplinary actions, including failing grades and school suspensions. More than half of the attackers (n = 21, 60%)
had both academic and disciplinary issues. For seven of the attackers (20%), a disciplinary issue at school was the
most recent stressor experienced prior to the attack.
GENERAL PERSONAL: Two-thirds of the attackers (n = 22, 63%) experienced stressors that were unique or personal in
nature, and therefore not as easily categorized. These stressors included a failed drug test, homelessness, and
feeling unhappy with one’s physical appearance. At least three attackers (9%) had witnessed violence against a friend
or loved one, including a death by stabbing.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
33
CHANGING SCHOOLS: Over half of the attackers (n = 19, 54%) had changed schools at least once, not including the
normal transition from primary to secondary schools. Those 19 attackers changed schools more than two times on
average. Seven attackers (20%) had switched schools during the six months leading up to the attack. The reasons
for the school changes included expulsions, placements in an alternative school setting, moving to a new residence,
enrolling in school after homeschooling, and transferring for personal reasons (e.g., to avoid bullies).
Some attackers experienced additional types of stressors that were less common, including stressors related to
criminal or judicial activity (n = 10, 29%), physical health problems (n = 9, 26%), and problems at work (n = 6, 17%).
Only eight attackers were known to have been employed, and most of them were employed in the food service industry.
The number of contexts in which the attackers in this study experienced stressors ranged from two to eight, with an
average of five. Two of the attackers experienced stressors in all eight contexts.
Bullying
Bullying is a problem faced by students and continues to be identied as a factor in NTAC's research on targeted
violence in schools. Most of the attackers in this study (n = 28, 80%) were bullied by their classmates. For more
than half of the attackers (n = 20, 57%), the bullying appeared to be of a persistent pattern which lasted for weeks,
months, or years.
Bullying includes unwanted, aggressive behavior among school-aged children with an intent to do physical, social, or
emotional harm; which involves a real or perceived power imbalance; and is, or could be, repeated.
26
Between
one-quarter and one-third of students in the United States report that they have been bullied, while almost
three-quarters of students and school ofcials have reported witnessing bullying.
27
The U.S. Department of Health and
Human Services warns of lasting negative effects for those who are bullied, those who bully others, and those who
witness bullying.
The evidence presented here supports past research ndings that bullying causes signicant stress and harmful
outcomes for students. Schools should adopt and enforce a zero-tolerance policy toward bullying and should
encourage students to report bullying when they witness it or are victims of it. School ofcials and other authorities
must act to intervene and stop bullying when it is reported to them. Bullying does not have to meet a legal threshold
before schools can intervene.
TYPES OF BULLYING: The Centers for Disease Control and Prevention (CDC) identies four types of bullying: verbal,
physical, social, and property.
28
This study additionally incorporated cyber bullying.
• Verbal: For nearly three-quarters of the attackers (n = 26, 74%), there was evidence that they experienced some
form of verbal bullying. This included spoken or written forms of aggression. Examples of verbal bullying experienced
by attackers prior to their attacks included threats, crude gestures, name-calling, teasing, taunting, and suggestions
that they kill themselves.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
34
• Physical: Many attackers (n = 14, 40%) experienced bullying that
involved physical aggression. Examples included hitting, kicking,
tripping, pinching, pushing, threatening or menacing gestures, and
nonconsensual sexual contact.
• Social: Six of the attackers (17%) experienced some kind of
social bullying, which involved words or actions designed to harm
the victim’s reputation or social standing. Examples included
having rumors spread about them, being ostracized by others
(e.g., classmates got up and left when the attacker sat down at a
table), and being maliciously ignored.
• Property: Four of the attackers (11%) experienced bullying involving acts of theft or destruction of property.
Examples included having money stolen and having toilet paper thrown at their house.
• Cyber: A few attackers (n = 3, 9%) experienced cyber bullying, which involved the use of digital technologies, such
as text messaging and social media platforms. Cyber bullying can be done in public or in private. Some cyber
bullying occurs in a way that may only be known to the victim (e.g., direct text messages), while other cyber bullying
may be broadcast for public view (e.g., publicly viewable social media posts). The attackers who were cyber bullied
were harassed on social media and received threatening emails, social media messages, and text messages.
Though verbal and physical bullying were the most common types of bullying experienced by the attackers, several of
the attackers experienced three or four different types of bullying.
VERBAL, PHYSICAL, AND PROPERTY BULLYING: A 12-year-old student fatally shot one teacher and wounded two classmates
before killing himself. The attacker had been teased and called names by his middle school classmates, including “gay,” “lazy,”
“stupid,” and an “idiot.” Classmates had poured water down his pants and then said the attacker had wet himself. In other
instances, the attacker had his pants pulled down by classmates, was tripped in the hall, and had money stolen from him. The
attacker referenced some of these incidents in one of the suicide notes he left, which was addressed to classmates
and teachers.
VERBAL, PHYSICAL, SOCIAL, AND CYBER BULLYING: A 13-year-old student stabbed and injured a classmate multiple times
in her head and arm at their middle school. The victim and her friends had called the attacker “ugly,” pushed her in the
hallway, spread rumors about her, and subjected her to threatening phone calls and Internet messages.
WHO KNEW ABOUT THE BULLYING: From case to case, there were differences regarding who was aware of the bullying
experienced by the attacker. In almost half of the cases (n = 16, 46%), at least one of the attacker’s parents was
aware that their child experienced bullying. In 16 cases (46%) a classmate was aware, and in 12 cases (34%) a school
ofcial was aware. In ve cases (14%), there was evidence that all three groups knew that bullying was taking place.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
35
A 13-year-old student attacked
classmates with a knife, injuring four.
Classmates knew that the attacker had
experienced bullying for years and that he
had threatened revenge. The attacker’s
parents were aware of the bullying. The
attacker had sought help from the
school nurse and guidance counselor, but
the situation did not improve.
In three cases (9%), the attacker perceived himself as a victim of bullying, but no one else shared that perception. For
example, one attacker noted in his journal that he had been bullied, yet no evidence could be identied to indicate the
attacker’s parents, classmates, or school ofcials were aware of any bullying or the attacker’s perceived victimization.
Of note, there were 10 instances (29%) where evidence indicates at least one classmate knew the attacker was being
bullied, but no school ofcials were aware. This highlights the need for schools to establish climates that encourage
both students and parents to share concerns with school staff. Providing students with a means to anonymously report
bullying, while also ensuring that all students have at least one positive relationship with a school staff member, will
help to encourage students to share actionable information when they observe incidents of bullying at school.
WHERE BULLYING OCCURRED: The attackers experienced bullying most frequently at school, including on the bus.
However, bullying was also experienced in private residences, in public spaces, online, and via phone calls and
text messages.
BULLYING TOLERATED: In a few cases, there was evidence that school ofcials knew about the bullying an attacker was
experiencing, but there was no indication that the school responded to address the bullying in any way. In other cases,
when school ofcials did respond to address the bullying, the responses were minor and ineffective, or the responses
made the bullying worse.
3
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
36
A 16-year-old student shot and injured his classmate and a teacher. Prior to the attack, the attacker had sought help from
a teacher after being bullied. In response, the teacher told the attacker to “man up.” The attacker’s mother later called the
school about the bullying, but there is no evidence that school ofcials took any action to intervene. The only intervention
occurred after the attacker got into a stght with his bullies, after which the attacker was suspended. At his trial after the
shooting, several classmates testied that the attacker had been relentlessly bullied and teased.
ATTACKERS AS BULLIES: Just over one-third of the attackers (n = 13, 37%) were bullies themselves. Prior to their
attacks, these attackers bullied their classmates verbally, physically, and through cyber bullying. Nearly all of the
attackers who had bullied others were also victims of bullying.
A 14-year-old student shot and injured one classmate in his high school cafeteria. The attacker had been teased and called
names. He had been struck with binders, punched, and pushed around. The attacker also had a history of bullying others.
He would insult other students or scream at them, and had pushed, kicked, and slapped his classmates.
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/ 2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
37
School Interactions & Relationships
STUDENT-TEACHER RELATIONSHIPS: Over one-third of the attackers (n = 13, 37%) had at least one positive
relationship with a staff member at their current school. Positive relationships were perceived from the student’s
and/or teacher’s perspective, and were evidenced by the attacker being well-liked by teachers; a teacher writing a job
recommendation for the attacker; teachers complimenting the attacker’s academic work; and teachers describing the
attacker as a model student, a good kid, or a pleasure to have in class. In addition, there were a few cases where
a school staff member attempted to make a positive impact on an attacker’s life prior to the attack, for example, by
looking out for a student or checking in with them on a weekly basis to make sure they were okay.
A similar number of attackers (n = 10, 29%) had a negative relationship with at least one staff member at their
current school. Negative relationships were characterized by the attacker previously disliking or threatening a teacher,
not feeling supported by staff, not receiving assistance from school staff when they needed help, thinking teachers
were disrespectful, and the attacker being disrespectful toward teachers or refusing to follow directions. Only three
attackers (9%) felt they had a neutral relationship with the staff at their current school, which was evidenced by the
attacker describing their relationship with teachers as cordial.
SOCIAL LIFE: The majority of attackers (n = 29, 83%) had at least one friend, and at least four of the attackers (11%)
were in a romantic relationship at the time of the attack. Only a few attackers (n = 5, 14%) appeared to have no
friends or romantic relationships.
29
It is important to note that these relationships were seen on a continuum. For
example, an attacker may have been described as having friends, but others also noted that the same attacker kept
to himself.
Academic Performance & Extracurricular Activities
Recent Academic Performance: Attackers’ recent academic achievement was categorized as a positive, neutral, or
negative performance for the school year in which they were enrolled at the time of the attack. For example, if the
attack took place in February 2012, the academic performance was captured for the 2011-2012 school year. If the
attack took place close to the start of the school year, the attacker’s academic performance was captured for the
previous school year. The academic performance of seven attackers could not be identied.
• Positive Performance: About one-third of the attackers (n = 12, 34%) demonstrated a positive academic
performance through one or more achievements, to include making the honor roll, earning grades of A’s and B’s, or
being enrolled in an Advanced Placement (AP) class or a gifted/talented program.
• Neutral Performance: Five attackers (14%) demonstrated a neutral academic performance. Those who were rated
as neutral included those who received a broader range of grades (e.g., A’s, B’s, and C’s) and attackers who were
able to raise their failing grades into more average levels.
• Negative Performance: About one-third of the attackers (n = 11, 31%) demonstrated a negative academic
performance, evidenced by failing classes, repeating a grade, not completing homework, or performing below
grade-level.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
38
ACADEMIC ASSISTANCE: Ten attackers (29%) received some form of academic assistance during their school career.
This assistance took the form of special education services, tutoring received or recommended, or working with an
English as a Second Language (ESL) teacher or a learning strategies teacher. Eight students had received an
Individualized Education Program (IEP) or a 504 plan.
CHANGES IN PERFORMANCE: Several attackers (n = 7, 20%) displayed a negative academic change in the year leading
up to their attack, which was most often displayed through declining grades. Due to the change, they also experienced
negative effects in other areas of their academics, for example, failing a class, being removed from an honors or gifted
program, or being removed from a school sports team. For example, one attacker had ended his junior year with a 3.0
GPA (B average). Within two months of the attack, the student's GPA declined to a 1.7, as he was earning D’s and F’s
in all of his classes during his senior year.
EXTRACURRICULAR ACTIVITIES: Nearly three-quarters of the attackers (n = 25, 71%) had been involved in an
extracurricular activity at school, in the community, or both. School extracurricular activities included sports, JROTC,
band, theater, orchestra, choir, debate team, art club, and Spanish club. Extracurricular activities in the community
included sports, Boy Scouts, lifeguarding, volunteering, chess club, and religious activities.
NOTEWORTHY ACHIEVEMENTS: Just under half of the attackers (n = 16, 46%) accomplished at least one noteworthy
achievement in their school career or in the community. These accomplishments were often related to:
• Academics: Attackers received honors or awards for academic achievement, were enrolled in Advanced Placement
(AP) courses or an honors/gifted/talented program, were selected to skip a class or grade level due to high
prociency, and were on the honor roll.
• Extracurricular Activities: Examples of extracurricular achievements by attackers include winning rst place on a
school sports team, receiving a scholarship for music camp, winning an essay contest, and winning a school
talent show.
• Social Achievements: An attacker was elected by classmates as Homecoming Prince.
• Other Personal Achievements: Examples of other personal achievements by attackers include achieving a weight
loss goal, becoming a deacon, earning a black belt in karate, appearing on television, gaining a large number of
subscribers to a YouTube channel, and successfully completing a marksmanship qualication program.
Disciplinary History
Almost three-quarters of the attackers (n = 25, 71%) received some form of school disciplinary action in the five
years prior to the attack. The behaviors that elicited disciplinary actions occurred along a continuum of severity,
ranging from cheating or using profanity to threatening or engaging in physical assaults. As with many of the topics
addressed in this report, the presence of these behaviors alone does not necessarily indicate that a student will carry
out a targeted school attack. However, gathering information about a student’s disciplinary history, along with other
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
39
behaviors and circumstances, will assist school ofcials in identifying appropriate interventions to support the student
and decrease the risk of harmful outcomes.
BEHAVIORS RECEIVING DISCIPLINE: Disciplinary actions resulted from a wide range of behaviors in the school setting.
Attackers were disciplined for the following types of behavior:
• Fighting: Fourteen attackers (40%) were disciplined for ghting at school. These incidents included ghts on the
school bus and hitting other students.
• Classroom Conduct: Twelve attackers (34%) were disciplined for behavior that affected the classroom environment,
including interrupting the teacher, causing a disturbance, using a cell phone, demonstrating a bad attitude, and not
following directions.
• Threatening/Violent Behavior: Eleven attackers (31%) were disciplined for displaying threatening or violent behaviors
at school. This included having weapons on campus; angry outbursts; using violent language; threatening a teacher,
the school, or classmates; drawing pictures that depicted killings; and sharing a note with classmates that implied
the attacker was going to do something bad.
• Academic Integrity: Seven attackers (20%) were disciplined for behaviors related to their academics, including
cheating on a test, failing to complete homework, and stealing another student’s paper and claiming it as their own.
• Inappropriate Language: Six attackers (17%) were disciplined for using inappropriate language in school. This was
not limited to profanity, but also included the use of vulgar or abusive language, inappropriate comments, and
name-calling towards teachers or classmates.
• Tardiness/Truancy: Five attackers (14%) were disciplined for being tardy to school or to class. Eight attackers (23%)
were disciplined for truancy, which included skipping class, leaving class and not returning, leaving school without
permission, and refusing to attend school.
• Banned Substances: Four attackers (11%) were disciplined for behaviors involving banned substances at school,
including selling and distributing marijuana, possessing cigarettes or a cigarette lighter, possessing drug
paraphernalia, and being under the inuence of alcohol or drugs.
• Harassment/Bullying: Four attackers (11%) were disciplined for bullying and harassing behaviors, which included
groping a female classmate and pulling down a classmate’s pants.
• Other Discipline: Beyond the categories described in this section, attackers were also disciplined for gambling,
vandalism, riding a scooter on school property, violating the school dress code, riding a different school bus without
permission, and miscellaneous misconduct incidents.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
CONSEQUENCES RECEIVED: The types of school discipline received as a result of these behaviors varied, and some of
the attackers received multiple actions for the same repeated behaviors. Twenty-ve attackers (71%) received
disciplinary action in the ve years prior to their attack. The level or severity of the discipline ranged from classroom
consequences (e.g., having the student’s seat moved) to being expelled or arrested, and there were a few cases in
which an attacker was disciplined, but the type of discipline received was not specied in the attacker’s school
records. Attackers received the following forms of discipline:
• Suspension: Eighteen attackers (51%) had received at least one suspension from school. Suspensions included
formal in-school suspensions (ISS) and out-of-school suspensions (OSS), as well as instances where a student was
sent home from school, even if the school did not classify the discipline as a suspension. Attackers were issued
suspensions for a broad range of behaviors, from disruptive behavior to death threats. The length of suspensions
also varied, from a half-day to multiple days.
A 16-year-old student shot and injured a classmate and a teacher in his high school science classroom. The attacker had been
previously suspended for ve days after making threats about killing other students on the school bus.
• Parental Contact: Thirteen attackers (37%) received discipline in the form of parental notication. Parents were
contacted through letters sent home, documents requiring a parent signature, phone calls home, and parent
meetings with school staff. Some, but not all, of the meetings with parents involved the attacker as well.
• Conversations with Staff: Eleven attackers (31%) received discipline in the form of conversations with teachers,
administrators, or SROs. These meetings either were one-on-one or occurred in a group setting. In one case, an
attacker had a reentry meeting with school staff to discuss his behavior and plan for his return to school following
a suspension.
40
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
41
• Charges or Arrests: Eight attackers (23%) were arrested or charged at school for various infractions, including
possession of marijuana, simple assault, disrupting public school, affray, trespassing, and disturbing the peace.
• Referrals for Evaluation/Assessment/Support Services: Eight attackers (23%) were referred to a team or mental
health professional for an evaluation, an assessment, or support services. These included referrals for a threat
assessment, risk assessment, or mental health evaluation. Referrals were also made to school counselors and
school psychologists, and some students were placed on behavioral contracts or behavioral intervention plans.
• Formal Documentation: Eight attackers (23%) received discipline in the form of formal documentation in their
school record. Schools used formal documentation to create a record in the form of behavior referrals, discipline
referrals, school hearings, and progress reports.
• Detention: Eight attackers (23%) received school detention or lunch detention, or were required to attend Saturday
school for their behavior.
• Classroom Consequences: Seven attackers (20%) received discipline in the form of classroom consequences.
These consequences included placing the attacker in timeout, modifying assignment instructions, conscating a cell
phone, moving the attacker’s seat, removing the attacker from the class temporarily, and transferring the attacker to
a different class permanently.
• Expulsion: Six attackers (17%) had been expelled from school. An expulsion included being removed from the
school system for a predetermined number of days or permanently removed from the school district. Expulsions
resulted from an accumulation of suspensions, ghting, threatening harm to other students or self, and bringing a
weapon to school.
• Other Consequences: Seven attackers (20%) received other school disciplinary actions, including removal from the
football team, having to write a statement about their behavior, and having a scooter conscated.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
42
Threat Assessment
A multidisciplinary school threat assessment team provides schools with a structured approach to identify students
exhibiting threatening or concerning behavior, gather information to assess if the student poses a risk of harm to
him/herself or the school community, and manage the risk through appropriate interventions, resources, and
supports.
30
A threat assessment should focus on a range of behaviors, from lower-level concerns (e.g., self-harm,
depressed mood, bullying) to imminent or direct threats of violence. It is critical that student threat assessments be
multidisciplinary and focused on providing robust interventions to students assessed as posing any level of risk to
themselves or others.
Four attackers in this study (11%) were referred to their school’s threat assessment team prior to the attack. In each
of these four cases, a student or teacher at the school reported to someone in a position of authority that the
attacker had made a direct threat toward classmates or school staff.
Who was the
threat directed
toward?
Case 1
Case 2
Case 3
Staff member
Specic student
Specic
student/school
Threat 1 -
students
Threat 2 -
students
Teacher Yes
Yes
No
Yes
Yes
The threatened
student
Teachers and 15
students
The threatened
student
The threatened
student
Within one year
prior to attack
Assessed as a low level of
risk; parents monitored
student; student saw
psychiatrist once a week
Assessed at risk level 4
out of 5, where 1 is the
highest risk;
behavior/academic
contract
Staff determined the
student did not meet
criteria to be assessed
Assessed as posing a
threat; expelled for 21
days; returned to school
for the start of the new
school year
Within one year
prior to attack
Within one year
prior to attack
Out-of-school suspension;
intervention plan
Within three years
prior to attack
Within two years
prior to attack
Case 4
Who reported
the threat?
Was a threat
assessment
completed?
When was the
assessment
completed?
What was the outcome?
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
43
Information was identied regarding some of the gaps and limitations in how the threat assessments in these four
cases were conducted. For example, an independent review of one case (Case 1) was requested by the school
superintendent and the local board of education following the attack. The resulting public report identied several
limitations in how the attacker’s threat assessment had been conducted.
31
According to the report, these
gaps included:
• “…data sources used for the threat assessments were not specied on the form and many forms lacked
specic behavioral information. Forms varied in completion and therefore it was also sometimes difcult to see
the reasoning behind assigned risk levels.”
• “All steps of the [school district’s] process as outlined by [the school district] do not appear to have been
followed by the [school] staff in the threat assessment of [the attacker].”
• “There was only a two-person team in the threat assessment of [the attacker]. The SRO was not involved at
the [threat assessment] meeting.”
• “No interviews were conducted with [the attacker] or his family outside the meeting process.”
Available court records related to Case 3 describe a threat assessment process that appeared comprehensive in
regards to the volume of information gathered by the team about the concerning student. Of the 15 students who
reported the attacker’s threatening behavior on the school bus, 7 led written incident reports to the school. As a
result, the attacker was suspended for ve days, and a threat assessment was initiated. The attacker met with the
assistant principal and the school psychologist, and the threat assessment team conducted interviews with the
attacker and his family, classmates, and teachers. The SRO checked the student’s residence for weapons, where
the ofcer only located paintball guns. The assessment revealed that the attacker had spoken about his “dream of
killing students,” was a loner, and had concerned the assistant principal so much that “she made a point to check on
him regularly and asked how he was doing almost daily.” Despite this thorough and appropriate information-gathering
process, the student was deemed to be low risk. The school rated him a “4” on a 5-point scale, where “1” would have
indicated high risk.
Concerning Behaviors
Every attacker included in this analysis (n = 35, 100%) exhibited concerning behaviors prior to their attack. In all
but two of these cases (n = 33, 94%), concerning behaviors were displayed at school. Three-quarters of the attackers
(n = 27, 77%) displayed concerning behaviors at home or in the community, and about three-quarters displayed them
online (n = 26, 74%).
Concerning behaviors ranged from relatively minor activities to actions that elicited fear in those who observed them.
For example, some attackers made statements that were simply out of character for the attacker or displayed other
minor changes in behavior, while in other cases, attackers made direct threats of violence or brought weapons to
school. In most of the cases (n = 28, 80%), the attacker’s behavior elicited concern from bystanders regarding the
safety of the attacker or those around them.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
44
These ndings support previous Secret Service research projects examining targeted violence in other contexts (e.g.,
mass attacks in public spaces
32
), which routinely nd that the vast majority of attackers displayed behaviors that
elicit concern in others prior to their attack. These behaviors provide opportunities for school staff, law enforcement,
families, and friends to recognize students who are in crisis and intervene as appropriate. In this report, concerning
behaviors include actions an attacker took, communications an attacker made, and other observable acts displayed by
an attacker prior to the attack.
Range in Severity
In most cases, attackers displayed a behavior that was so concerning that it should have been met with an immediate
response, including a threat assessment. These behaviors are seen as being objectively concerning or prohibited.
Examples include threats to cause harm, violent acts, bringing weapons to school, and suicidal statements. Of note,
references to suicide, though not necessarily an indication of potential violence, are concerning and require immediate
intervention to protect the student involved. Thirty-one attackers (89%) had displayed objectively concerning or
prohibited behaviors that were observed by others. These behaviors were reported to the attackers’ parents (n = 15,
43%), school staff (n = 14, 40%), and other ofcials (n = 8, 23%), including law enforcement. In two-thirds of the
cases (n = 23, 66%), at least one prohibited behavior was observed by classmates or adults and was not reported.
Most attackers also displayed a range of lower-level concerning behaviors that observers may not have recognized as
concerning without knowing the full context of the situation. These behaviors are seen as being part of a constellation
of lower-level behaviors, and may not warrant an immediate safety response. Nonetheless, they should elicit some
level of concern. Examples of these behaviors include a depressed or angry mood, conicts between classmates, and
an interest in violent topics. When these types of behaviors are assessed in context with other factors, the level of
concern could increase. Thirty attackers (86%) exhibited a constellation of such lower-level behaviors.
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
45
Common Themes
The analysis revealed 10 common themes among the concerning behaviors that were displayed by the attackers,
meaning those behaviors that were observable to others. For example, while 28 attackers had experienced bullying,
only 12 openly spoke about it. Each of the attackers in this study exhibited between two and eight of the below listed
themes, with an average number of five.
• Threats to the Target or Others, and/or Intent to Attack: Most of the attackers (n = 29, 83%) shared verbal, written,
visual, or video communications that referenced their intent to carry out an attack, threatened the target, and/or
threatened others.
• Intense or Escalating Anger: Three-quarters of the attackers (n = 26, 74%) displayed behaviors or shared
communications indicating signicant or increasing anger. These behaviors included having angry outbursts at
school and engaging in aggressive acts at home.
• Interest in Weapons: Nearly three-quarters of the attackers (n = 25, 71%)
stockpiled weapons or communicated about weapons in a way that
indicated an unusual or concerning level of interest. For some, their
interest in weapons was expressed through drawings or artwork, while
others had developed a reputation among classmates for having
an intense interest in guns, knives, or explosives. A few attackers had
built and detonated explosives.
• Sadness, Depression, or Isolation: Nearly two-thirds of the attackers
(n = 22, 63%) either spoke about their sadness, depression, or loneliness,
or appeared through their observable behaviors to be experiencing these
feelings. Some attackers conded in others about their feelings or wrote
about them online or in school assignments. Bystanders also observed the
attackers isolating themselves, withdrawing from others, appearing sad,
or crying.
• Changes in Behavior or Appearance: More than half of the attackers
(n = 20, 57%) exhibited observable changes in demeanor, appearance,
or routine prior to the attack. Examples included increased apathy about
life, decreased effort on school work, withdrawing from activities,
decreased personal hygiene, spending time in new places or with new
people, changes in eating and sleeping patterns, and changes in online
behaviors (e.g., changing prole picture to something unusual for
the person).
Common Themes of
Concerning Behaviors
Threats to the target or
others, and/or intent to attack
Intense or escalating anger
Interest in weapons
Sadness, depression,
or isolation
Changes in behavior
or appearance
Suicide and/or self-harm
Interest in violence
Talk of being bullied
Concerns over
grades/attendance
Harassing others
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
46
• Suicide and/or Self-harm: Half of the attackers (n = 19, 54%) had communicated about, or engaged in behaviors
related to, suicide or self-harm. In some cases, multiple friends knew that the attacker was suicidal. In other
cases, evidence of self-harm was noted by friends, parents, and/or school staff.
• Interest in Violence: About one-third of attackers (n = 13, 37%) spoke or wrote about their violent interests,
including topics related to previous school attacks, Hitler/Nazism, and other violent themes. Sometimes these
interests were shared openly, while in other instances the attacker was more subtle (e.g., viewing photos of
previous school attacks while in the cafeteria).
• Other common themes among the concerning behaviors included attackers talking about being bullied (n = 12,
34%), poor grades or attendance that elicited concerns from parents or school staff (n = 10, 29%), and the
attackers harassing others (n = 5, 14%). In three of the ve cases involving harassment, the victim of the
harassment was later targeted in the attack.
Communications
Among the concerning behaviors displayed by the attackers, the
communications they made to others provide particular insight
into their thinking, motivations, and intentions. The majority of
attackers shared concerning communications verbally through
in-person statements (n = 31, 89%). About half shared
concerning electronic messages sent to a specic person or
persons (n = 20, 57%) or posted online to groups or the general
public (n = 17, 49%). One-third of the attackers (n = 12, 34%)
conveyed concerning thoughts through school assignments (e.g.,
writing essays on violent topics, drawing weapons on binders
or papers).
A 13-year-old student injured four classmates using a knife. Though disarmed by peers, he was later found to have also
brought a blowtorch, gasoline container, water bottle lled with a clear ammable liquid, recrackers, a bottle of vodka, a
mask used to prevent the inhalation of fumes, and several other knives. About one week prior, the attacker posted a video
on YouTube titled, “Testing my Blowtorch,” in which he lit the ame on a gas blowtorch, presumably the same one he brought
during the attack.
Timing of Concerning Behaviors
Three-quarters of the attackers (n = 26, 74%) displayed their most recent concerning behaviors within the two days
prior to their attack. Nearly all of the attackers (n = 32, 91%) shared or exhibited such behaviors within one month of
their attack. It should be noted, however, that these behaviors were not necessarily new, as nearly half of the
attackers (n = 14, 40%) had been displaying concerning behaviors for more than a year prior to their attack.
Conveyance of
Communication
In-person
Online posting
School assignments
Phone
Electronic message
to specic person(s)
31
17
12
1
89%
49%
34%
3%
20 57%
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
47
Most attackers (n = 27, 77%) threatened their targets or shared their intentions to carry out an attack. In two-thirds of
the cases (n = 23, 66%), those communications gave some level of imminence to the attack. About 90% of attackers
who selected specic targets had made a threat prior to the attack (see Appendix A: Statistical Analyses).
A 15-year-old student fatally shot a classmate in her high school. The night before, the attacker texted a friend stating, “I can’t
take it anymore. I want to kill myself so bad, the pain she causes hurts too bad. I just want to blow my brains out.” The next
morning, she texted a friend at another high school saying that she was going to shoot someone and go to jail. Within an hour
before the shooting, the attacker told a friend in class that she was going to make the victim feel “pain,” showed the classmate
her handgun, and began crying.
Final Communications: Half of the attackers (n = 19, 54%) prepared or delivered nal communications prior to the
attack. While some attackers kept these communications to themselves (e.g., left a suicide note in their bedroom),
over one-third of the attackers (n = 13, 37%) sent these communications to others. Final communications included
manifestos, journal entries, goodbye messages to friends and family (sent by text message, email, or Internet posts),
and wills or funeral requests.
A 15-year-old student fatally shot four of his classmates and injured one other in his high school cafeteria before killing
himself. All of the victims were his close friends or cousins. The day prior, he tweeted, “It won’t last…It’ll never last” and texted
his ex-girlfriend asking her to meet after school so he could say goodbye. He also sent her a Facebook message that said,
"Read my messages tomorrow from 7:15 to 12:30.” The attacker posted on Instagram, “Tell my mom I love her,” and included
an emoji image of a gun. On the day of the attack, the attacker drafted a message on his phone titled, "My funeral shit,” which
outlined how his funeral should be handled. Minutes before the shooting, he sent the prepared text message about his funeral
to 14 of his family members. He also texted his father, telling him, “Read the paper on my bed. Dad. I love you.”
Three of these attackers also took other actions that were akin to nal acts, including one attacker who deleted his
email account, one who hugged her father goodbye when being dropped off at school (which she had never done
before), and another who said goodbye to his pets.
Response to Concerning Student Behaviors
The response elicited from those individuals who observed the attackers’ concerning behavior varied. The more
cautious responses included actively avoiding the attacker (n = 7, 20%) or expressing concern to other peers
(n = 8, 23%). In two-thirds of the attacks (n = 23, 66%), someone approached the attacker at least once to ask if the
attacker was okay, offer help, or attempt to dissuade the attacker from engaging in any harmful acts toward
Timing of Threats
Two-thirds of the attackers (n = 23, 66%) communicated their intent to attack or threatened the target within two
weeks of the attack, and half (n = 19, 54%) did so within two days. Over one-third of the attackers (n = 13, 37%)
communicated their intent or threatened the target within an hour of the attack.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
48
self or others. In fewer than half of the incidents (n = 14, 40%), adults received information about what a bystander
had observed. These included classmates telling a parent or school staff member (n = 8, 23%) and adults talking
to other adults to discuss their concerns about the attacker (n = 14, 40%). Ten cases (29%) involved other types of
responses, including:
• Parents: searched rooms for drugs, set a new curfew, conscated an attacker’s pellet gun or knife, gave their son
a punching bag to help him deal with anger, led reports of their concern to the school district, asked the school to
develop an IEP for their child;
• School Staff: removed the student from an activity, made the student throw away violent drawings, sent a memo to
other school staff informing them that the attacker and the victim were not to be left alone together, developed
plans to escape the school if the student were to do something violent, considered resignation from the school to
protect self and the community;
• Students: warned peers of the attacker’s intentions if they were viewed as potential targets, led reports of their
concern to the school district.
In just under half of these cases (n = 16, 46%), the observed behavior elicited a significant response, including school
staff calling the police in response to a criminal act or concern for safety, school staff suspending or expelling the
attacker, and family members getting the attacker a mental health evaluation or therapy. In one case, the attacker
himself asked for mental health treatment after police responded to a physical altercation between him and his
mother at home.
In two-thirds of the attacks (n = 23, 66%), there was at least one communication by the attacker about his or her
intent to attack, or another observed threatening behavior, to which there was not a response. Some of the
reasons provided by bystanders for not reporting their concerns included:
• Classmates believed that the attacker was joking, or they dismissed the threat, because the attacker had a
history of making threatening statements and had never done anything before.
• Classmates believed that they had helped calm the attacker down enough that they would not act on the
suicidal or homicidal statements, or classmates had checked in on the attacker and decided he or she was ne.
• Classmates feared what the attacker may do to them if they reported what they knew.
• Classmates intended to try to help the attacker with his or her problems, but the attack occurred sooner than
they expected.
• Classmates believed the attacker would do something violent, but did not think it would happen at school.
• A teacher assumed that an administrator would take action because the administrator had observed the same
concerning behavior that the teacher had witnessed.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
49
PART III: IMPLICATIONS
Like the February 14, 2018, shooting at Marjory Stoneman Douglas
High School in Parkland, FL, this analysis suggests that many of
the 41 attacks described in this report could have been prevented.
Like the Parkland attacker, many of the attackers in this study had a
history of violent, threatening, and other concerning behavior; prior
contact with law enforcement; instability in the home; access to and
inappropriate interest in weapons; or issues related to psychological,
emotional, or behavioral factors. While every situation is unique and
should be treated as such, one common factor across all of these
tragedies is that there appears to have been an opportunity to
identify and intervene with the attacker before violence occurred.
Threat assessment is the best practice for preventing targeted violence. The goal of a threat assessment is to
identify students of concern, assess their risk for engaging in violence or other harmful activities, and deliver
intervention strategies to manage that risk. Schools should implement a threat assessment process in conjunction
with the most appropriate physical security measures as determined by the school and its community.
Threat assessment procedures should recognize that concerning student behaviors occur along a continuum, from a
constellation of lower-level concerning behaviors (e.g., depressed mood and behavior changes) to behaviors that are
objectively concerning or prohibited (e.g., threats of harm and physical assaults). Many of these behaviors that elicit
concern may not involve physical violence or criminal acts, but still require an assessment and appropriate
intervention. The threshold for intervention should be low, so that schools can identify students in distress before
their behavior escalates to the level of eliciting concerns about safety.
The threat assessment process begins with establishing a comprehensive targeted violence prevention plan, as
described in Enhancing School Safety Using a Threat Assessment Model: An Operational Guide for Preventing
Targeted School Violence. The guide provides a framework for establishing an effective threat assessment process
intended to proactively intervene with students who may pose a risk of violence or other unwanted behavior. The steps
include establishing a multidisciplinary threat assessment team of highly trained school personnel; dening
concerning and prohibited behaviors; establishing and providing training on a central reporting mechanism;
determining the threshold for law enforcement intervention; establishing threat assessment procedures and
investigation protocols; developing risk management options; creating and promoting a safe school climate; and
providing training for school personnel, students, parents, and law enforcement.
The ndings in this report underscore the importance of establishing a targeted violence prevention plan. A
multidisciplinary threat assessment team, in conjunction with the appropriate policies, tools, and training, is the best
practice for identifying, assessing, and managing students who elicit concern. Major ndings, and their implications on
the threat assessment process, are highlighted here.
States Mandating Threat Assessment
In 2013, Virginia became the first state to
legislatively mandate the use of threat
assessment teams in K-12 schools under
state law.
In more recent years, other states have
passed similar laws that require threat
assessment teams at the school or district
level. Some additional states have proposed
such legislation.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
50
Key Findings and Implications
There is no profile of a student attacker, nor is there a profile for the type of school that has been targeted.
Attackers varied in age, gender, race, grade level, academic performance, and social characteristics. The use of a
student prole or checklist does not yield accurate or useful results when determining the risk a student poses for
conducting a school attack. Similarly, there was no identied prole of the type of school impacted by targeted
violence, as schools varied in size, location, and teacher-to-pupil ratio. Rather than focusing on a set of traits or
characteristics, a threat assessment process should focus on gathering relevant information about a student’s
behaviors, situational factors, and circumstances to assess the risk of violence or other harmful outcomes.
Attackers usually had multiple motives, the most common involving a grievance with classmates.
Discovering a student’s motive for engaging in concerning behavior is critical to assessing the student’s risk of
violence, and it allows the threat assessment team to intervene and develop management strategies that can redirect
the student away from violent choices. This nding also highlights the importance of providing training on conict
mediation and coping skills for students. When multiple motives are involved, the school may need to implement
multiple strategies to reduce the risk of unwanted behavior.
Most attackers used firearms, and firearms were most often acquired from the home.
The ndings of this report illustrate that students can be very resourceful in accessing rearms stored in the home.
While many of the attackers used unsecured rearms, others were able to gain access to rearms that were secured
in a safe. It should be further noted, however, that some students perpetrated attacks using knives instead of
rearms. Therefore, a threat assessment should explore if a student has access to any weapons, with a particular
focus on weapons access at home. Schools, parents, and law enforcement must work together rapidly to restrict a
student’s access to weapons in those cases when a student poses a risk of harm to self or others.
Most attackers had experienced psychological, behavioral, or developmental symptoms.
A recently published study that surveyed a sample (n = 16,000) of 10th- to 12th-grade students found that providing
or increasing mental health services for students was the most common factor selected by students as something
that would increase feelings of safety at school, selected by more than one-third (38%) of students surveyed.
33
The
ndings of this survey, and those highlighted in this report, illustrate the importance of ensuring that students and
parents have access to, and are informed about, mental health resources, social services, and substance use
treatment. It should also be noted that many of the attackers in this study had received prior mental health treatment,
illustrating that mental health treatment should be viewed as a component of managing risk.
Half of the attackers had interests in violent topics.
Half of the attackers in this study had displayed an unusual or concerning interest in violence or weapons. These
types of interests, without an appropriate explanation, should initiate further information-gathering, assessment, and
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
51
management by school personnel. For example, a student who is preoccupied or xated on topics like the Columbine
shooting or Hitler, as was noted in several of these cases, may be the focus of a school threat assessment to
determine how such an interest originated and if the interest negatively impacts the student’s thinking and behavior.
All attackers experienced social stressors involving their relationships with peers and/or romantic partners.
Every attacker in this study experienced at least one social stressor, frequently related to bullying. Many attackers
experienced stressors in other aspects of their lives, as well, including family, academics, and school discipline. In
addition to adopting and enforcing zero-tolerance policies on bullying, schools should also provide students with
strategies related to stress management and the development of coping skills. All school personnel, including
teachers, administrators, and other staff members, should be trained to recognize signs of a student in crisis.
Additional training should focus on crisis intervention, deescalation, and suicide prevention.
Nearly every attacker experienced negative home life factors.
The negative home life factors experienced by many of the attackers included parental divorce or separation, drug
use, criminal charges among family members, and domestic abuse. While none of the factors included here should be
viewed as predictors that a student will be violent, past research has identied an association between many of these
types of factors and a range of negative outcomes for children. In some cases, school staff may not be aware of such
home life factors that may be impacting how a student behaves at the school. Information-sharing among agencies
is crucial in helping to address the impact of negative home experiences. For example, some cities and states have
enacted laws that require law enforcement or other emergency care providers to inform school ofcials if a student is
involved in, or present at, the scene of a traumatic event. Information-sharing among schools, law enforcement, social
services, and courts can help to ensure that the students most in need of additional resources do not
go unrecognized.
Most attackers were victims of bullying, which was often observed by others.
In many cases, the school was aware that the attacker had experienced bullying, but the responses to the bullying
varied. In some cases, the school did little to intervene or intervened in a way that caused the bullying to worsen.
Students should be encouraged to report bullying that they experience or observe, both in and out of school. Students
should also be provided an avenue to anonymously report such concerns, so that they can do so without fear of
reprisal from classmates. Further, it is critical that schools implement comprehensive programs designed to promote
safe and positive school climates, where students feel empowered to support the social and emotional wellbeing
of classmates.
Most attackers had a history of school disciplinary actions, and many had prior contact with law enforcement.
Most attackers had a history of receiving school disciplinary actions resulting from a broad range of inappropriate
behavior. The most serious of those actions included the attacker being suspended, expelled, or having law
enforcement interactions as a result of their behavior at school. An important point for school staff to consider is that
punitive measures are not preventative. For example, two of the attackers in this study were suspended from school
at the time of their attacks. If a student elicits concern or poses a risk of harm to him/herself or others, removing the
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
52
student from the school may not always be the safest option. In other instances, disciplinary actions will be deemed
necessary. To help in making the determination regarding appropriate discipline, schools should employ disciplinary
practices that ensure fairness, transparency with the student and family, and appropriate follow-up. For example, the school
should make efforts to maintain a positive rapport with the student and family, even during those instances of suspension
or expulsion. This might include checking in on the student during his or her time out of school and taking steps to ensure
a positive transition for the student when they return.
All attackers exhibited concerning behaviors. Most elicited concern from others, and most communicated their intent
to attack.
The initial indicators of a student who was in distress or exhibiting concerning behavior were often observed by peers,
school staff, family members, or others in their immediate community systems. The behaviors that elicited concern ranged
from a constellation of lower-level concerns to objectively concerning or prohibited behaviors. Most of the attackers
communicated a prior threat to their target or communicated their intentions to carry out an attack. In many of these cases,
someone observed a threatening communication or behavior but did not act, either out of fear, not believing the attacker,
misjudging the immediacy or location, or believing they had dissuaded the attacker. These ndings continue to highlight the
importance of encouraging students, school personnel, and family members to report troubling or concerning behaviors, in
order to ensure that those in positions of authority can intervene. These same community members need to be trained on
identifying risk factors for student violence and students in crisis.
Conclusion
NTAC’s experience delivering training and consultations for public safety entities has provided a rsthand account of what
works and of the challenges facing schools, law enforcement, and others tasked with keeping communities safe.
NTAC knows that many of the recommendations offered in this report, and in NTAC’s Enhancing School Safety Guide, can
be implemented by schools and their local partners right away. However, some key strategies for keeping schools safe
can best be implemented only with the appropriate resources, policies, and authorities to support the effort. For example,
several states have recently passed laws requiring schools or school districts to implement multidisciplinary school threat
assessment teams. Similarly, some states have passed laws to create statewide, anonymous reporting tools, following the
model of Safe2Tell
TM
Colorado.
In some cases, grant funding may be available to support these efforts. For example, the STOP School Violence Act of
2018 provides the U.S. Department of Justice’s Community Oriented Policing Services (COPS) Ofce the authority to
provide monetary awards directly to states, units of local government, and Indian tribes to improve security at schools
through evidence-based school safety programs, specically including school threat assessment programs and anonymous
reporting systems.
34
Some individual states have provided similarly intended grants to their schools and school districts
as well.
The safety of children in school is a topic that everyone agrees on. Ensuring their safety requires leadership and vision as
well as common sense. A thorough review of the ndings contained in this report should make clear that tangible steps can
be taken to reduce the likelihood that any student would cause harm, or be harmed, at school.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
53
CREATING A TARGETED VIOLENCE PREVENTION PLAN
In July 2018, the Secret Service National Threat Assessment Center (NTAC) released Enhancing School Safety
Using a Threat Assessment Model: An Operational Guide for Preventing Targeted School Violence, which outlined
eight actionable steps for implementing a comprehensive targeted violence prevention plan in schools. The
guide, which is available on the Secret Service public website, provides a framework for schools to identify,
assess, and manage students who pose a risk of violence or other harmful behavior.
Step 1: Establish a multidisciplinary threat assessment team of school personnel including faculty, staff,
administrators, coaches, and available school resource ofcers who will direct, manage, and document
the threat assessment process.
Step 2: Dene concerning behaviors, including those that are objectively concerning or prohibited, which should
trigger an immediate intervention (e.g., threats, violent acts, or weapons on campus), and other
lower-level concerning behaviors (e.g., depressed mood, interest in violent topics, or conicts
between classmates).
Step 3: Establish and provide training on a central reporting system, like a smartphone application, an online
form, or a dedicated school email address or phone number. Ensure that it provides anonymity to
those reporting concerns and is monitored by personnel who will follow up on all reports.
Step 4: Determine the threshold for law enforcement intervention especially if there is a risk of harm to self
or others.
Step 5: Establish threat assessment procedures that include practices for maintaining documentation,
identifying sources of information, reviewing records, and conducting interviews. The assessment
should be guided by an understanding of the thinking and behavior observed in past school
attackers, as described in Protecting America’s Schools: A U.S. Secret Service Analysis of Targeted
School Violence.
Step 6: Develop risk management options to enact, once an assessment is complete. Create individualized
management plans to mitigate identied risks. Notify law enforcement immediately if the student is
determined to pose an imminent risk of harm to self or others. Take steps to ensure the safety of
potential targets, create a situation less prone to violence, redirect the student’s motive, and reduce
the effect of stressors.
Step 7: Create and promote a safe school climate built on a culture of safety, respect, trust, and emotional
support for students. Encourage communication, intervene in conicts and bullying, and empower
students to share their concerns.
Step 8: Provide training for all stakeholders, including school personnel, students, parents, and
law enforcement
.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
54
FINDINGS TABLE
General Information (n = 41)
Investigative Themes (n = 35)
Backgrounds (n = 35)
Gender
History of weapons use
Violent interests
History of anger or aggression
Home life factors (top 3 of 8)
Stressors (top 3 of 8)
Concerning behaviors (top 5 of 10)
Substance use/abuse
Mental health symptoms
Mental health treatment
Bullied
83% male, 17% female
77%
49%
74%
94%
100%
100%
49%
91%
54%
80%
Disposition: charged
School disciplinary histories (top 3 of 12)
Arrests or criminal charges:
Other context:
Bullied others:
History of violent behavior:
Psychological:
Behavioral:
Neurological/Developmental:
Fighting:
Class misconduct:
Threatening/aggressive behaviors:
Separation/divorce of parents:
Finance:
Arrest/incarceration of family member(s):
Social:
Family:
Academic or school disciplinary issues:
Threats/intent to attack:
Anger:
Weapons-related:
Depression:
Behavior changes:
20 as adults, 10 as juveniles, 2 as both
71%
31%
29%
37%
51%
69%
57%
20%
40%
34%
31%
71%
69%
54%
100%
91%
89%
83%
74%
71%
63%
57%
Age: average/range
Contact with law enforcement
15 / 12-18
49%
deceased
7 suicides, 2 killed by law enforcement
Final communications or gestures
Suspended, expelled, or law enforcement intervention
Elicited concern about safety
54%
60%
80%
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
APPENDIX A: STATISTICAL ANALYSES
Statistical tests revealed some differences among the attackers on topics involving their amount of planning,
communication of threats, bullying experienced, target selection, motive, weapon used, and harm caused. Fisher’s exact
tests of independence and chi-square tests of independence evaluated whether there were relationships between two
dichotomous variables, with the appropriate test selected based on the number of cases within the subcategories (i.e.,
cross-tabulations with fewer than ve cases in a cell required Fisher’s exact test). Computation of t tests for independent
samples compared averages of some continuous variables (i.e., number of planning behaviors and number of deaths)
across subsamples. All signicance tests were two-tailed. The cutoff for statistical signicance was an alpha level of .05;
however, Appendix A presents some statistically non-signicant, but substantively meaningful, ndings. The results should
be interpreted with the understanding that students were only coded “yes” when there was evidence of the presence of the
variable, and that absence of evidence is not equivalent to evidence of absence of the variable.
Note: Due to the limited behavioral information available for 6 of the attackers, only 35 cases were included in
the statistical analysis for the following three tests.
• Although not statistically significant, attackers who engaged in more planning behaviors caused harm to a greater
number of victims: Mass attackers averaged more than twice as many planning behaviors as attackers with fewer victims
[t (33) = 1.98, p = .06].
• Most attackers who targeted specific people had made a prior threat: About 90% of attackers who selected
specific targets had made a threat prior to the attack (p = .01). Less than 60% of attackers who targeted random
individuals had made a threat in advance (p = .04).
.
Table 2: Threatening Communications and Target Selection
Threat (n = 27)
No threat (n = 8)
Total
Random Target Selection
(p = .04)
No
n (col.%)
No
n (col.%)
Yes
n (col.%)
Yes
n (col.%)
19 (90%)
2 (10%)
21 (100%)
3 (38%)
5 (63%)
8 (100%)
8 (57%)
6 (43%)
14 (100%)
24 (89%)
3 (11%)
27 (100%)
Specic Target Selection
(p = .01)
55
Table 1: Number of Victims and Average Number of
Planning Behaviors
Number of Victims
One or two victims (n = 25)
Three or more victims (n = 10)
Average
1.4
3.1
SD
2.1
2.8
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
56
• Although not statistically significant, a larger share of attackers who experienced a persistent pattern of bullying
had bullying as a motive for carrying out the attack: Of the 20 attackers who experienced a persistent pattern of
bullying, 12 (60%) had bullying as a motive, compared with one-third who were not persistently bullied
(
2
= 2.44, p = .12)
Note: All 41 cases were included in the statistical analysis for the following tests.
• Nearly all attackers primarily motivated by a desire to kill targeted random victims: Six of the seven attackers (86%)
whose primary motive was a desire to kill targeted randomly (p < .01). Only two of the seven attackers (29%) primarily
motivated by a desire to kill had specic targets, compared with most individuals with a different motive (n = 29, 85%;
p < .01). Five of the ten attackers (50%) who only targeted random victims were primarily motivated by a desire to kill
(p < .01).
• Incidents in which attackers used firearms were significantly more fatal than those involving knives: While the
majority of both types of attacks were nonfatal, nearly half of the rearm attacks (n = 12, 48%) resulted in one or
more deaths, while only 2 of 16 knife attacks (13%) were fatal (p = .04).
Table 3: Persistent Bullying and Bullying as a Motive
Bullying as a motive
(n = 17)
Attack not motivated by
bullying (n = 18)
Total
Persistently Bullied
5 (33%)12 (60%)
10 (67%)
15 (100%)
8 (40%)
20 (100%)
Not Persistently Bullied
Table 4: Target Selection and Desire to Kill
Random Target Selection
(p < .01)
Specic Target Selection
(p < .01)
Yes (n = 16)
Yes (n = 31)
No (n = 25)
No (n = 10)
Total
Total
Desire to Kill was
Primary Motive
6 (86%)
2 (29%)
1 (14%)
5 (71%)
7 (100%)
7 (100%)
10 (29%)
29 (85%)
24 (71%)
5 (15%)
34 (100%)
34 (100%)
Other Primary
Motive
Table 5: Fatalities and Weapon Used
Fatalities
Zero
Two
Total
One
Four
Three
Mean (SD)
Firearm
13
0
25
10
1
1
0.68 (0.99)
Blade
14
0
16
2
0
0
0.13 (0.34)
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
57
APPENDIX B: COMPARISON TO THE SAFE SCHOOL INITIATIVE
In 2002, the U.S. Secret Service’s National Threat Assessment Center (NTAC) released the Safe School Initiative (SSI),
which examined 37 incidents of targeted violence in schools committed by 41 attackers between 1974 and June 2000.
Released 17 years later, Protecting America’s Schools examines 41 incidents committed by 41 attackers between 2008 and
2017. In examining the major ndings from both studies, a number of similarities were identied. In both studies, there
was no one prole of a school attacker and there were similar percentages of those with criminal histories, substance
abuse, types of mental health symptoms, and histories of being bullied or bullying others. Additionally, the majority had
experienced stressors, exhibited concerning behaviors, and made their intentions to attack known to others. In Protecting
America’s Schools, however, the analysis delved deeper and identied additional factors that were not considered as part of
the original study, including, but not limited to, the types of stressors, ways in which the attackers were bullied, the breadth
of issues related to home life, the behaviors that elicited disciplinary action and their consequences, and various aspects
of school dynamics.
The following table offers some of the statistics from the two studies; however, comparisons between the two should be
made only with caution as there were variations in research design and methodology, including variations of denitions,
availability of case information, and access to Internet-powered research tools. For instance, 17 years ago, studying
targeted violence in K-12 schools was new and identifying incidents would have been somewhat more challenging as there
were no databases of incidents or lists of attacks to consult. Further, in more recent years, more attention has been
focused on incidents of targeted violence, especially those affecting schools. Therefore, it is no surprise that similar
numbers of incidents were found even though the timespan covered in the present study (10 years) is less than half that of
the SSI (26 years). These factors may have also impacted why the current study identied more attacks than the SSI that
were perpetrated using bladed weapons.
Given these caveats, the following table presents the statistics which are most comparable.
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
58
93% 100%
THE INCIDENTS
Years
THE ATTACKERS
SCHOOL PERFORMANCE
BEHAVIORS
n = 37
1974 - 2000
n = 41
n = 41
2008 - 2017
n = 35 (of 41)
Safe School Initiative
(2002)
Protecting America's Schools
(2019)
Multiple attackers
Victims that were specifically targeted
Committed suicide
History of any arrest/charge
Subject was abused/neglected
Mental health diagnosis before
Substance use and/or abuse
Suspended (at least once)
Concerning behaviors
*Reects calculations as reported in the Final Report and Findings of the Safe School Initiative (2002).
At least one person knew
Perceived as loners
Expelled (at least once)
History of violence
Bullied by other students
Stressors
Subject bullied other students
Academic performance
a peer
an adult
Mental health symptoms
violent crime
8%
46%
13%*
27%
27%
17%*
49%
27%
83%
34%
10%
31%*
71%
98%
34%
83%
7%
17%
0%
56%
17%
31%
23%
40%
49%
51%
77%
26%
17%
51%
80%
100%
37%
77%
14%
17%
Weapons
Handguns
Long guns
Firearm from a residence
Depression
Suicidal thoughts or gestures
Suicide attempts
Positive
Neutral
Negative
Not found
Gender, age, status
97% firearm; 3% bladed;
8% had additional weapons
67% of 36 firearms attacks
50% of 36 firearms attacks
73% of 36 firearms attacks
61%
78%
10%
41%
37%
5%
17%
100% male; ages 11-21
95% current students
61% firearms; 39% bladed;
7% had additional weapons
72% of 25 firearms attacks
36% of 25 firearms attacks
76% of 25 firearms attacks
63%
63%
11%
34%
14%
31%
20%
83% male; ages 12-18
90% current students
United States Secret Service
NATIONAL THREAT ASSESSMENT CENTER
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
59
1
National Threat Assessment Center. (2018). Enhancing school safety using a threat assessment model: An operational guide for preventing targeted school violence. U.S.
Secret Service, Department of Homeland Security.
2
It should be noted that homicides rarely occur at schools in the United States. During the 2015–2016 school year, the most recent year for which data are available, about
1% of school-aged youth homicides in the United States occurred in a school setting; found in Musu, L., Zhang, A., Wang, K., Zhang, J., and Oudekerk, B.A. (2019). Indicators
of school crime and safety: 2018. Washington, DC: U.S. Department of Education and U.S. Department of Justice. Retrieved October 19, 2019, from
https://nces.ed.gov/pubs2019/2019047.pdf.
3
Vossekuil, B., Fein, R., Reddy, M., Borum, R., & Modzeleski, W. (2002). The final report and findings of the Safe School Initiative: Implications for the prevention of school
attacks in the United States. Washington, DC: U.S. Secret Service and U.S. Department of Education.
4
A former student was dened as a student who left the targeted school within the previous year and no longer attended a K-12 school, or a student who was currently enrolled
in any K-12 school and targeted a school he or she had previously attended. For example, while the perpetrator of the 2012 attack at an elementary school in Newtown, CT
was a former student, that attack was not included in the data for this report as the perpetrator had not attended a K-12 school in over three years prior to his attack.
5
National Center for Education Statistics. (2019). CCD public school data 2017-2018, 2018-2019 school years. Retrieved October 18, 2019, from https://nces.ed.gov/ccd/.
6
Public School Review. (2019). Average public school student:teacher ratio. Retrieved October 18, 2019, from
https://www.publicschoolreview.com/average-student-teacher-ratio-stats/national-data.
7
National Center for Education Statistics. (2019). CCD public school data 2017-2018, 2018-2019 school years. Retrieved October 18, 2019, from https://nces.ed.gov/ccd/.
8
Centers for Disease Control and Prevention. (2018). 1992–2016 School-Associated Violent Death Surveillance System (SAVD-SS) (partially funded by the U.S. Department of
Education, Ofce of Safe and Healthy Students). Unpublished tabulation. Retrieved October 19, 2019, from https://nces.ed.gov/programs/crimeindicators/ind_01.asp.
9
One attack did not have clear information available on the incident duration and was categorized as “unknown.”
10
One school-aged victim was not enrolled as a student at the school where the attack occurred. She was enrolled in a different high school and was attending prom at the
targeted school at the time of the attack.
11
The status of one of the attackers is unknown.
12
National Center for Injury Prevention and Control. (2017). 10 leading causes of death by age group, United States - 2017. Retrieved October 19, 2019, from www.cdc.gov:
https://www.cdc.gov/injury/images/lc-charts/leading_causes_of_death_by_age_group_2017_1100w850h.jpg.
13
For one incident that involved planning behaviors, the timespan of the planning behaviors was not available.
14
Giffords Law Center to Prevent Gun Violence. (2018). Minimum age to purchase and possess. Retrieved October 19, 2019, from https://lawcenter.giffords.org/gun-laws/
policy-areas/who-can-have-a-gun/minimum-age/#federal.
15
Centers for Disease Control and Prevention. (2019). Many children lack access to mental health care. Retrieved October 19, 2019, from
https://www.cdc.gov/childrensmentalhealth/documents/access-infographic.html.
16
One of the attackers with suicidal ideations was not included in the psychological symptoms category due to a lack of additional symptoms indicative of a psychological or
emotional health disorder. However, the attacker did experience a number of behavioral symptoms and he was appropriately included in that category.
17
Kjelsberg, E. (2006). Exploring the link between conduct disorder in adolescence and personality disorders in adulthood. Psychiatric Times, 23(8). Retrieved October 19, 2019,
from https://www.psychiatrictimes.com/exploring-link-between-conduct-disorder-adolescence-and-personality-disorders-adulthood.
18
Felitti, V.J, Anda, R.F., Nordenberg, D., Williamson, D.F., Spitz, A.M., Edwards, V., Koss, M.P., & Marks, J.S. (1998). Relationship of childhood abuse and household
dysfunction to many of the leading causes of death in adults. The Adverse Childhood Experiences (ACE) Study. American Journal of Preventative Medicine, 14(4):245-58.
19
Moore, K., Sacks, V., Bandy, T., & Murphey, D. (2014). Fact sheet: Adverse childhood experiences and the well-being of adolescents. Retrieved October 18, 2019, from
https://www.childtrends.org/wp-content/uploads/2014/07/Fact-sheet-adverse-childhood-experiences_FINAL.pdf.
20
Dallaire, D.H., & Wilson, L.C. (2010). The relation of exposure to parental criminal activity, arrest, and sentencing to children’s maladjustment. Journal of Child and Family
Studies, 19, 404-418.
21
Smith, V.C., & Wilson, C.R. (2016). Families affected by parental substance use. Pediatrics, 138(2).
22
Grant, K.E., Compas B.E., Stuhlmacher A.F., Thurm A.E., McMahon S.D., & Halpert J.A. (2003). Stressors and child and adolescent psychopathology: Moving from
markers to mechanisms of risk. Psychological Bulletin, 129(3), 447-66.
23
Shonkoff, J.P., & Garner, A.S. (2012). The lifelong effects of early childhood adversity and toxic stress. Pediatrics, 129, e232-e246.
24
Miller, A.J. (n.d.). Stress as a factor in family violence. Retrieved October 18, 2019, from https://www.unicef-irc.org/article/984-stress-as-a-factor-in-family-violence.html.
25
Sontag, L.M., Graber, J.A., Brooks-Gunn, J., & Warren, M.P. (2008). Coping with social stress: Implications for psychopathology in young adolescent girls. Journal of Abnormal
Child Psychology, 36(8), 1159–1174.
26
This denition was adapted from the denitions found in: U.S. Department of Health and Human Services. (2019). What is bullying? Retrieved October 19, 2019, from
https://www.stopbullying.gov/what-is-bullying/index.html; Gladden, R.M., Vivolo-Kantor, A.M., Hamburger, M.E., & Lumpkin, C.D. (2014). Bullying surveillance among youths:
Uniform definitions for public health and recommended data elements (Version 1.0). Retrieved October 19, 2019, from https://www.cdc.gov.
27
U.S. Department of Health and Human Services. (2019). Facts about bullying. Retrieved October 19, 2019, from https://www.stopbullying.gov/media/facts/index.html.
28
These terms were adapted from the denitions found in: Gladden, R.M., Vivolo-Kantor, A.M., Hamburger, M.E., & Lumpkin, C.D. (2014). Bullying surveillance among youths:
Uniform definitions for public health and recommended data elements (Version 1.0). Retrieved October 19, 2019, from https://www.cdc.gov.
29
For one attacker, there was insufcient information to assess their social life.
30
National Threat Assessment Center. (2018). Enhancing school safety using a threat assessment model: An operational guide for preventing targeted school violence. U.S.
Secret Service, Department of Homeland Security.
31
Kanan, L., Nicoletti, J., Garrido, S., & Dvoskina, M. (2016). A review of psychological safety and threat assessment issues related to the shooting at [high school name] on
December 13, 2013. Retrieved October 19, 2019, from https://www.safeandsoundschools.org/additional-resources-threat-assessment/.
32
National Threat Assessment Center (2019). Mass attacks in public spaces - 2018. U.S. Secret Service, Department of Homeland Security.
33
Croft M., Moore, R., Guffy, G. (2019). Creating safe schools: Examining student perceptions of their physical safety at school. Retrieved October 19, 2019, from
https://www.act.org/content/act/en/research/pdfs/R1767-school-safety-brief.html.
34
U.S. Department of Justice. (n.d.). School Violence Prevention Program (SVPP). Retrieved October 19, 2019, from https://cops.usdoj.gov/svpp.
Production photo credits
Cover image courtesy: Bill Chizek/Flag/Getty, Table of contents image courtesy: Code6d/Topics/Getty, School bus image courtesy: Edward Shackleford/Bus/Getty,
Student lunch image courtesy: Kali9/Concepts/Getty, Classroom image courtesy: Skynesher/School Building/Getty, Police line image courtesy: Carlballou/Manufactured Object/Getty,
Lady justice image courtesy: Querbeet/Fictional Character/Getty, Student at school entrance image courtesy: Monkeybusiness Images/Educational/Getty, Bullying image courtesy: FatCamera/Harassment/Getty,
Troubled student image courtesy: Izusek/Young Adult/Getty, Student detention image courtesy: LumiNola/Ethnicity/Getty, Troubled student at lockers image courtesy: Josh Blake/Human Age/Getty
Revised December 3, 2019
PROTECTING AMERICA’S SCHOOLS/ANALYSIS OF TARGETED SCHOOL VIOLENCE/2019
National Threat Assessment Center
