2D1

Learning from Error in Violence Prevention: A School Shooting as an Organizational Accident

Sarah Goodrum1 , Jessie Slepicka2 , William Woodward1, and Beverly Kingston1

Abstract

This article argues that the organizational structure and culture of schools may impede the prevention of violence in America’s schools, specifically threat assessment and management for students of concern. The data come from a qualitative case study of a school shooting where two students died; the data include deposition testimony from 12 school officials and more than 4,000 pages of school and law enforcement records. The findings illustrate the way the school’s organizational structure and culture shaped and hin- dered violence prevention practices. The tightly coupled guidelines for threat assessment created an insti- tutional myth of safety and a false sense of security for the school and district, and the loosely coupled structure of the organization led educators to modify guidelines and make decisions about the student’s behavior problems and discipline without consulting others. The school’s culture of autonomy for staff and fresh start mentality for students created unintentional secrets about the history of the student’s difficul- ties, which gave educators little context for understanding the problem behaviors they observed and inhibited the threat assessment team’s ability to adequately evaluate and monitor those behaviors. Recom- mendations for building organizational structures and cultures that support violence prevention in schools are discussed.

Keywords

school organization, school policy, violence in schools, at-risk students, tracking

School administrators, law enforcement officials,

and federal agencies have increasingly taken

active measures to prevent targeted acts of school

violence in the United States. Over the past 20

years, more than half of middle and high schools

in the U.S. have adopted an anonymous tip line

(Planty et al. 2020), and the U.S. Secret Service

has twice issued federal guidelines on threat

assessment in schools (Fein et al. 2004; National

Threat Assessment Center [NTAC] 2018). From

2014 to 2017, the National Institute of Justice

(2020) awarded approximately $246 million in

grants to support nearly 100 projects on school

safety. Despite these prevention efforts, there

were 57 active shooter incidents in U.S. schools

from 2000 to 2018 (Blair and Schweit 2014; Fed-

eral Bureau of Investigation 2016, 2018, 2019). In

2020–2021, more school shootings with casualties

1University of Colorado Boulder, Boulder, CO, USA 2The Pennsylvania State University, State College, PA,

USA

Corresponding Author:

Sarah Goodrum, Center for the Study and Prevention of

Violence, Institute of Behavioral Science, University of

Colorado Boulder, 1440 15th Street, Boulder, CO

80309, USA.

Email: [email protected]

Sociology of Education 2022, Vol. 95(4) 257–275

� American Sociological Association 2022 DOI: 10.1177/00380407221120431

journals.sagepub.com/home/soe

Research Article

occurred in the United States than in any other

academic year since data collection began (Irwin

et al. 2022); as of July 2022, 27 school shootings

have resulted in injury or death in the United

States this academic year (‘‘School Shootings

This Year’’ 2022). The investigations that follow

these tragic events consistently identify ‘‘failures

of foresight’’ to recognize and interrupt the attack-

er’s path toward violence within schools and

across communities (Erickson 2001; Fein et al.

2004; Goodrum et al. 2018, 2019; Goodrum,

Woodward, and Thompson 2017; Marjory Stone-

man Douglas High School Public Safety Commis-

sion [MSDHSPSC] 2019; NTAC 2019; Turner

1976; Virginia Tech Review Panel 2007). Recent

reports on the Uvalde school shooting at Robb

Elementary School, where the attacker killed 19

children and two teachers, show a similar trajec-

tory (Burrows, Moody, and Guzman 2022; Con-

don 2022).

Studies from sociology and organizational psy-

chology suggest these failures of foresight arise

when an organization’s structure creates a culture

that inhibits employees from recognizing, evaluat-

ing, and addressing problems or concerns (Doyle

2010; Fox and Harding 2005; Vaughan 2016). In

a detailed study of two school attacks, Newman

and colleagues (2004) found that the loosely cou-

pled structure of the schools discouraged staff

from seeing and interpreting students’ pain and

rage as needing intervention. Building on this

work, Fox and Harding (2005) concluded that

the schools lacked the resources to help staff rec-

ognize and address students’ social and mental

health needs. In 2002, the U.S. Secret Service

and Department of Education released the Safe

School Initiative, a study of 37 school attacks

(Vossekuil et al. 2002). The Initiative informed

the development of threat assessment guidelines

to help school staff identify and evaluate students

of concern (Fein et al. 2004; NTAC 2018; Ran-

dazzo et al. 2006; Reeves and Brock 2018; Vosse-

kuil et al. 2002).

Although school attacks are still rare in the

United States, the number of school attacks with

fatalities rose from 5 in 2009–10 to 27 in 2019–

20 (Irwin et al. 2021), raising questions about

our approach to violence prevention. Costa

(2012) argues that when complex problems persist

even after the introduction of straightforward sol-

utions, we need to reexamine the organizational

structure, culture, and beliefs shaping those prob-

lems and forestalling the solutions. Admittedly,

the availability of firearms is a key societal factor

that accounts for high homicide rates in the United

States compared to other Western countries (Win-

temute 2015). Thus, individuals intent on using

a firearm for criminal purposes are likely to find

a way to obtain one (Alper and Glaze 2019;

Thrush 2021). Acknowledging this reality, this

article focuses on processes that can avert and pre-

vent violence and that have been proven to save

lives (Langman and Straub 2019). Using case

study evidence from a high school attack where

two students died, this research examines the

role of the school’s organizational structure on

the evaluation and management of the attacker in

the weeks and months leading up to his deadly

rampage. This study builds on the literature in

sociology, organizational theory, and violence pre-

vention in general and the work of Newman and

colleagues (2004), Fox and Harding (2005), and

Vaughan (2016) in particular.

ORGANIZATIONAL ACCIDENTS

Tragic Errors

Sociologists and psychologists have argued that

major organizational accidents, such as wrongful

convictions, aviation disasters, and medical errors,

are not the result of one mistake or one bad apple

employee; instead, they are the result of a series of

smaller mistakes that build on each other and com-

pound to a tragic end (Dörner 1996; Doyle 2010;

Vaughan 2016). In these cases, the structure of

the organization fosters a culture that leads

employees to let small problems go unaddressed,

sometimes repeatedly (Reason 1998; Vaughan

2016). In the preface to the second edition of her

book The Challenger Launch Decision, Vaughan

(2016) described her and others’ dismay to learn

that the same organizational problems she identi-

fied in 1996 with NASA’s Challenger disaster

(1986) contributed to NASA’s Columbia disaster

(2003). We feel similar dismay to learn about

another targeted school attack in the United States.

Research on targeted school attacks repeatedly

finds that a ‘‘patchwork of information mini-

systems’’ hindered school and university officials

from identifying and discussing students’ concern-

ing behaviors (Erickson 2001; MSDHSPSC 2019;

Newman et al. 2004; Virginia Tech Review Panel

2007; see Doyle 2010:124). Prior to their attacks,

many attackers became disengaged from school,

258 Sociology of Education 95(4)

experienced difficulties with school or law

enforcement, and were recognized as troubled

(Vossekuil et al. 2002; see also NTAC 2019).

Connecting these pieces of information together

to heed the warning signs, however, remains chal-

lenging because the problem behavior often arises

across different settings and with different person-

nel (Virginia Tech Review Panel 2007). The solu-

tion to connecting these pieces of information

together appeared to be threat assessment and

management procedures for students of concern.

In 2004, the U.S. Secret Service and Depart-

ment of Education’s federal guidelines on threat

assessment and management sought to help

schools evaluate the likelihood a concerning stu-

dent would perpetrate an act of targeted violence

(Fein et al. 2004; see also NTAC 2018). The

guidelines created a framework for the threat

assessment process, and they were meant to

address the challenges that Newman, Fox, Har-

ding, and others identified in a school’s ability to

address a student’s behavioral and mental health

issues. States, safety centers, and school districts

across the country have used the guidelines to

develop school-based threat assessment policies,

protocols, and procedures. Organizational theo-

rists, however, would note that formal guidelines

rarely match actual work activity, creating an

‘‘institutional myth’’ (Meyer and Rowan 1977).

This case study examines the match between

guidelines and activities in the threat assessment

process for one student of concern and how that

match influenced decision-making.

The Coupling Continuum

The level of match between an organization’s formal

bureaucratic guidelines and the staff’s implementa-

tion of those guidelines is called coupling, and the

level of match can range from loosely to tightly cou-

pled (Meyer and Rowan 1978; Paino 2018). In

a loosely coupled organization, the parts are mini-

mally interdependent of each other for everyday

functioning such that a problem in one part does

not typically disrupt work in another part (Bidwell

1965; Fox and Harding 2005; Meyer and Rowan

1977, 1978; Paino 2018). In a tightly coupled orga-

nization, the match between policies and procedures

is strong and is maintained through formal bureau-

cratic controls and leaders’ close oversight (Bidwell

1965; Meyer and Rowan 1978; Paino 2018).

Schools are frequently characterized as the

quintessential example of a loosely coupled

system due to educators’ autonomy over curricu-

lum delivery and classroom management.

Recently, however, researchers have provided

a more nuanced explanation for the organizational

structure of schools. Paino (2018) argues that

schools have three levels of coupling (not just

one):(1) micro-level (i.e., principal–teacher rela-

tionships), (2) meso-level (i.e., district–school

relationships), and (3) macro-level (i.e., govern-

ment policies–school relationships). In addition,

decoupling describes the disconnect between

bureaucratic rules and work activity, and recou-

pling describes the effort to reset an organization

from loosely to tightly coupled so that bureau-

cratic rules align with work activity and uphold

‘‘institutional mythical’’ ideals (Hallett 2010).

Other researchers have noted that one organization

can support both loosely and tightly coupled struc-

tures, and all types of coupling have their positives

and negatives.

Examples of the coupling continuum and the

upsides and downsides of coupling emerge in all

types of organizations, including schools. As an

example of tight coupling, state and federal guide-

lines provide strict rules for schools on the fre-

quency and reporting of fire drills, student atten-

dance, and academic testing. These rules provide

clear guideposts for action, but they can also hin-

der creativity and autonomy. Examples of loose

coupling include teachers’ freedom to manage

their classrooms and student conduct with minimal

oversight from administrators, which can benefit

job satisfaction and limit communication (Bidwell

1965; Paino 2018). Sociologists have found that

unintentional silos of information can arise when

there is little need for communication across parts

of a loosely coupled system (Doyle 2010; Fox and

Harding 2005).

Structural Secrecy

Fox and Harding (2005) refer to the information

lost within loosely coupled organizations as

‘‘structural secrecy’’ because critically important

details about a person or case remain unknown

to other decision makers (see also Newman et al.

2004; Vaughan 2016). In these organizations, the

division of labor and hierarchical structure make

communication less necessary and also more diffi-

cult. Interestingly, what starts as a highly efficient

organizational structure can create bureaucratic

and human obstacles to learning about and

Goodrum et al. 259

relaying information from one employee to

another and from one office to another (Doyle

2010; Fox and Harding 2005; Newman et al.

2004; Vaughan 2016; Weber 1978). For example,

schools have access to a plethora of student infor-

mation (e.g., academic records, testing scores,

family status, immunization records), but they do

not always have information about students’

behavioral, mental health, or criminal histories

(Erickson 2001; Fox and Harding 2005; Goodrum

et al. 2017; MSDHSPSC 2019; Vaughan 2016;

Virginia Tech Review Panel 2007). The concern-

ing behaviors addressed in a teacher’s classroom

or documented in a teacher’s email to parents

may not carry forward when that student moves

from one classroom, grade level, or school to the

next (MSDHPSC 2019; Virginia Tech Review

Panel 2007).

The public health approach to violence preven-

tion recommends that schools and communities

work to ‘‘identify and assess youth from a very

young age . . . [to deliver] effective mental health

and educational service[s], and [facilitate] . . .

cross-system communication among professionals

charged with the care of children’’ (Office of the

Child Advocate 2014:3; see also Kingston et al.

2018). This type of ‘‘cross-system communica-

tion,’’ however, may prove difficult to operation-

alize in complex educational settings, where the

organizational structure (e.g., procedures, policies,

staffing, resources) does not include consistent or

institutionalized support for violence prevention

and intervention. In addition, existing procedures

and policies may discourage staff from sharing

information about students in crisis. Some

research suggests that misinterpretations of the

Family Educational Rights and Privacy Act

(FERPA) can lead school officials to withhold

information about students’ concerning behavior

from other staff or agencies (Chapman 2009;

Goodrum et al. 2017). These structural realities

can create gaps in violence prevention in schools,

but we have yet to fully understand how these

structural realities shape cultural norms and every-

day practices for school safety.

Using the sociological and psychological liter-

ature on organizations to examine the evidence

from a targeted school attack, this qualitative

case study asks the following questions: (1) How

was the school’s organizational structure coupled?

(2) How did the school’s organizational structure

influence staff communication and decision-

making? (3) Did the school’s organizational

structure inhibit violence prevention? If so, how

might staff develop strategies to overcome the

structural order of schools, and between districts,

to prevent violence and support troubled students?

By illustrating the specific ways a school’s organi-

zational structure shapes communication and deci-

sions about students of concern, we identify the

path toward safer and more aware school systems.

DATA AND METHODS

Data Collection

Data for the study came from an investigative arbi-

tration on the circumstances leading to a school

attack where two students died (Goodrum et al.

2017, 2018, 2019). To date, the case remains one

of only two known cases where a threat assess-

ment was conducted with a student prior to their

deadly attack (for another case, see MSDHPSC

2019). The investigative arbitration yielded

numerous interviews and thousands of documents,

including (1) deposition testimony from 12 school

and district officials covering more than 2,500

transcript pages, (2) 27 PDF documents containing

more than 4,000 pages of text from the sheriff’s

office’s investigation, (3) 171 PDF documents

containing more than 4,200 pages of text from

the district, and (4) 64 exhibits produced and intro-

duced during deposition testimony. The 12 school

and district officials deposed included nine school

staff (i.e., principal, two assistant principals,

school psychologist, two teachers, counselor,

school resource officer, campus security officer)

and three district administrators. Nine of the depo-

nents were male, three were female, and all were

Caucasian.

The depositions offer detailed narrative

accounts of the circumstances, interactions, and

communications prior to the attack. Similar to

qualitative interviews, the deposition testimony

allowed participants to describe their perceptions

and experiences in their own words. Each deposi-

tion took between two and eight hours. To empha-

size the lessons learned instead of personal identi-

ties, all participants are identified by their job title

and a number when more than one participant held

the title, such as Assistant Principal 1, Assistant

Principal 2, District Official 1, School Psycholo-

gist, and School Resource Officer. To avoid con-

tributing to the subculture of violence, the pseudo-

nym ‘‘JD’’ is used in place of the attacker’s name.

260 Sociology of Education 95(4)

The project data provide rich insight on the

school shooting as an organizational accident for

two reasons. First, other research on school vio-

lence consistently finds that people knew some-

thing was wrong with the attacker prior to their

attack, but they did not know who to tell or how

to intervene (Erickson 2001; MSDHSPSC 2019;

Virginia Tech Review Panel 2007). By examining

educators’ decision-making process, we start to

understand the organization’s management of

and response to a troubled student. Second, as

the frontline workers often blamed for these tragic

events, educators are intimately knowledgeable

about students and schools and are best situated

to reflect on and offer lessons learned on threat

assessment and school safety. These staff provide

detailed information on the context for school pro-

cedures and decision-making, which can inform

the development of effective intervention strate-

gies (Harding, Fox, and Mehta 2002; Yin 2014).

Case study method. The qualitative case

study method provides a rich and meaningful

investigation of the structural factors that influ-

enced the situation and decision-making (Birkin-

shaw, Brannen, and Tung 2011; Bryman and

Bell 2011; George and McKeown 1985; Gephart

2004; Mahoney 1999, 2000; Maxwell 2013).

This method provides an in-depth, multifaceted,

and contextualized examination of the phenome-

non and detailed insight into an uncommon form

of behavior, a targeted school attack (Crowe

et al. 2011; Geis 1991; Harding et al. 2002;

McCutcheon and Meredith 1993; Yin 2014).

As an extensively utilized research design in

sociology, case studies allow researchers to build

a knowledge base for how to evaluate abstract

theoretical concepts and principles (e.g., cou-

pling, structural secrecy) to inform professional

practice and scientific understanding (Crowe

et al. 2011; Flyvbjerg 2006; Gill and Stenlund

2006; Keen and Packwood 1995; McCutcheon

and Meredith 1993; Yin 2014). More specifi-

cally, the case study provides knowledge that

transcends the specific case to offer lessons

learned for other cases (Mills, Durepos, and

Wiebe 2010; Stake 1995, 2005). This approach

is particularly useful when answering what,

why, and how research questions (Crowe et al.

2011) and when trying to understand and prevent

acts of targeted school violence (Harding et al.

2002). To set the stage for the findings, we

provide some background information on the

school, the student, and his family.

Case background. In the fall of 2013, an 18-

year-old White male senior named JD threatened

to kill his debate coach, yelling ‘‘I’’m gonna kill

that guy’’ in the parking lot of the high school after

he was demoted from a leadership role in the

school’s debate club. Another teacher overheard

JD’s outburst and reported it to an assistant princi-

pal, who requested a threat assessment. The assis-

tant principal and a school psychologist formed

the high school’s two-person threat assessment

team, and they met with JD and his parents to ask

him questions about his threat, risk behaviors, and

support system. During the assessment, the team

noted that ‘‘Mom reports ‘deep seeded anger’’’

and that the student admitted he ‘‘had anger issues

for a while’’ (School Psychologist). The team eval-

uated him as a low-level risk for violence. Shortly

after the assessment and unbeknownst to the team,

however, the student began keeping a diary on his

tablet documenting his plan to perpetrate violence

at the school during finals week.

The postattack investigation revealed that JD

had a history of low-level behavior problems. In

elementary school, he hit students with his lunch

box, kicked a student in the stomach, and hit

another student in the head. JD’s student behavior

records document his claims of being ‘‘picked

on’’ (Exhibit 19); however, the investigation did

not reveal specific instances of victimization but

instead identified multiple instances where JD ver-

bally and physically bullied classmates (Exhibit

24). In the months following the threat assessment,

JD exhibited problem behaviors in several differ-

ent settings and had an enraged outburst in a for-

eign language class. (For a detailed timeline of

his behaviors, see Goodrum et al. 2017). Three

months after his threat assessment and two days

after his outburst, he entered the school with

a shotgun and other weapons; he was looking for

the debate coach, but he killed a classmate and

himself instead.

The large suburban high school where this

shooting happened included approximately 2,200

students and a little more than 100 full-time teach-

ers. Approximately 80 percent of students identify

as White, and 7 percent of students receive free or

reduced-priced lunch. The debate coach described

the school’s culture as embracing ‘‘perfection-

ism,’’ wishing that school staff and students could

Goodrum et al. 261

admit that ‘‘it’s not perfect here . . . mistakes are

made and [students] can learn from those mis-

takes. And we tell kids that all the time [but] I

wonder how true it is [in this school]’’ (Debate

Coach).

The attacker’s parents may have sensed the

school’s culture of perfectionism. JD’s parents

were not part of the arbitration agreement or

deposed in the case, but the evidence suggests

they had a strained relationship with the school.

For example, one teacher described JD’s mother

as nonresponsive to her emails about his profane

language and inappropriate comments in class

and said she arrived 75 minutes late to pick up

her son from a cross-country event (Teacher 1).

During a parent-teacher conference, this teacher

said JD’s father ‘‘sat down [with her] and . . .

said something like, ‘What horrible things do

you have to say about my son, like every teacher

in this place?’’’ (Teacher 1). During the threat

assessment, JD’s mother described him as having

anger issues, and she requested but did not provide

documentation to support an Individualized Edu-

cation Plan for her son (School Psychologist). As

is common in postattack investigations, we do

not have detailed information from the attacker’s

parents, which may be a limitation of the project

data (Fein et al. 2004; NTAC 2019).

Data Analysis

The deposition transcripts are the main data source

for the study. To code the transcripts, we relied on

Strauss’s (1987) guidelines for qualitative data

analysis and line-by-line coding. Each transcript

was read several times and coded using themes

from the literature on organizational theory (i.e.,

a deductive approach; Bradley, Curry, and Devers

2007). The codes included tight coupling, loose

coupling, and structural secrecy. Coding was exe-

cuted through a five-step process. First, for each

code, the first and second authors conducted

a line-by-line review of each deposition transcript.

Second, when an excerpt illustrated a theme, the

excerpt was highlighted and labeled with the

code, using the qualitative data management soft-

ware NVivo. This process allowed the coded text

to be organized into electronic folders by code

while retaining information about the deposed wit-

ness’s job title (e.g., Teacher 1). When an excerpt

addressed more than one theme, it was coded for

all relevant themes with a note about the cross-

listing. Third, once all transcripts were coded, all

of the text excerpts within each code were

reviewed to ensure correct placement and to iden-

tify subthemes. During this review, two new

themes emerged: the fresh start mentality and

autonomy. In the fourth step, all transcripts were

reviewed and coded for the newly identified

themes. Finally, to consider the possibility that

educators’ roles influenced their perceptions, we

reviewed and compared the coded data by role

(e.g., teacher vs. administrator). These compari-

sons reveal the competing interests (e.g., student

privacy, school safety) shaping educators’ deci-

sion-making. Here, we present the quotes and

case details that most clearly illustrate the findings

(Burnard et al. 2008). Additional project data from

the school’s records, law enforcement investiga-

tion, and deposition exhibits provide context for

deposition testimony.

FINDINGS

Organizational theory provides a powerful tool for

examining the circumstances leading to a tragic

accident. Schools, like other organizations, fall

on a coupling continuum from tight to loose and

can use recoupling to tighten up decoupled com-

ponents (Hallett 2010; Meyer and Rowan 1978;

Paino 2018). Recoupling allows organizations to

reestablish the connection between bureaucratic

ideals and actual work activity, but these efforts

can feel frustrating to staff (Hallett 2010). The

findings presented here illustrate the ways the

tightly and loosely coupled components of the

school and district influenced the implementation

of federal threat assessment guidelines, encour-

aged a culture of autonomous decision-making,

and created structural secrets about the extent

and seriousness of the student’s troubles.

Tightly Coupled Guidelines for Threat Assessment

On paper, the district’s threat assessment and man-

agement policies and procedures were tightly cou-

pled to federal guidelines, offering clear steps for

action with students of concern (Firestone 1984;

Weick 1976). For example, in 2004, approximately

nine years before the attack, the district started pro-

viding schools with information on threat assess-

ment, and by 2008, the district had created ‘‘stan-

dardized [district] threat assessment protocols . . .

[which] aligned with the recommended model

262 Sociology of Education 95(4)

from the U.S. Secret Service, the FBI, and Dr.

Smith [a pseudonym]’’ (Exhibit 9). These standards

included a two-hour threat assessment team train-

ing, a 51-slide threat assessment presentation, and

a four-page threat assessment and action plan pro-

tocol (Exhibits 9, 4, and 35, respectively). The dis-

trict’s threat assessment protocol closely modeled

the U.S. Secret Service’s guidelines, which is note-

worthy due to the state’s tragic history with targeted

attacks (Erickson 2001; Hamm 2021). In this con-

text, the Secret Service’s framework for threat

assessment provided legitimacy for the district’s

safety practices, offering a powerful example of

recoupling as a response to tragedy (Hallett 2010).

The tight match between the Secret Service’s

guidelines for threat assessment and the district’s

policies on threat assessment started to loosen

with implementation, specifically in the district’s

oversight of the school-level threat assessment

team and the team’s investigation of the student

(Goodrum et al. 2018, 2019). For the team, the dis-

trict’s Coordinator of Student Support Services

explained:

[T]here’s a lot of [federal] documentation

and guidance that says you need an estab-

lished [threat assessment] team. . . . But

[we do] not necessarily [think it needs to

be] the exact same three or four people

[for every threat assessment]. (District Offi-

cial 1)

This district official did not expect schools to

strictly follow federal guidelines for threat assess-

ment teams, creating a small buffer between the

bureaucratic rules and actual staff activities. In

this case, the team included two staff, not three

or four as recommended by federal guidelines

(Fein et al. 2004). This represented the start of

the decoupling between guidelines and action on

threat assessment.

Other research from this case indicates that the

implementation of the U.S. Secret Service’s six

principles and 11 questions for threat assessment

also differed from guidelines such that not all of

the principles or questions were addressed with

the student of concern (Goodrum et al. 2018,

2019). The Coordinator of Student Support Serv-

ices explained:

[We] don’t [spend] hours and hours in

a training and making sure [staff] know

exactly how to fill out every box [on the

threat assessment protocol]. The point of

the [district] training is: Can we help people

get those big principles?

This district official’s rational explanation for the

purpose of the threat assessment training captures

the ambivalence schools may feel when trying to

balance an outside entity’s formal guidelines

with their informal cultural practices. Certainly,

strict adherence to guidelines can create consis-

tency in procedures, but it may also depersonalize

interactions and escalate conflict (Meyer and

Rowan 1977, 1978). As one example, strict zero-

tolerance policies and exclusionary discipline

practices to crack down on students’ behavior

problems have been shown to disproportionately

target students of color (Porowski, O’Conner,

and Passa 2014). Loosely applying guidelines for

threat assessment, however, may leave uninten-

tional, and even dangerous, gaps in the school’s

knowledge about and understanding of a student’s

potential for violence.

This district’s reluctance to tightly couple

threat assessment guidelines with staff activities

is not unique. The field of implementation science

consistently finds that the delivery of prevention

programs proves challenging in school settings

(Fixsen et al. 2005; Greenberg 2010; Mihalic

and Irwin 2003). In fact, fewer than 50 percent

of school-focused prevention programs rely on

best-practice methods for implementation (Biglan

et al. 2003; Gottfredson and Gottfredson 2002;

Ringwalt et al. 2009; U.S. Department of Educa-

tion 2011). Translating guidelines into practice

proves difficult in busy school settings, where staff

are underresourced and overburdened (Biglan

et al. 2003; Gottfredson and Gottfredson 2002;

Osher 2012; Ringwalt et al. 2009). The district’s

Coordinator for Student Support Services

described principals’ frustration with strict threat

assessment guidelines:

A principal [will say to me,] ‘‘Everybody

[who is talking about a threat assessment]

safety plan says, ‘We need to have eye-to-

eye supervision on this kid,’ [but] I don’t

have the staff to do it.’’ . . . [Or they’ll

say to me,] ‘‘We want . . . this kid [to

meet] on a weekly basis with [the school]

psychologist,’’ but their schedule is so

booked, it makes it difficult. . . . These prin-

cipals want to know, ‘‘[Should] I imple-

ment the safety plan fully and take staff

Goodrum et al. 263

away from other pieces . . . or [should] I

say, ‘I can’t have this kid in my building.’’’

Schools simply lack the capacity and resources to

implement violence prevention programs and prac-

tices with fidelity (Forman et al. 2009; Mihalic

and Irwin 2003). Thus, providing strict mandates

and close supervision may seem like a solution for

an effective threat assessment process, but tight cou-

pling is likely to falter if the school’s organizational

structure, staffing, and resources cannot support

them. In this scenario, adopting the Secret Service’s

guidelines for threat assessment without the institu-

tional support to manage, support, and monitor a stu-

dent of concern creates an institutional myth of

school safety (Edelman et al. 2011; Hallett 2010;

Myer and Rowan 1978). Hallett (2010:42) explains,

‘‘myths are idealized cultural accounts, not neces-

sarily something false . . . and [they] provide legit-

imacy’’ but often do not match workers’ activity. In

this case, the institutional myth of threat assessment

provided a false sense of security that the right sys-

tems were in place to prevent a targeted school

attack.

Loosely Coupled Components and Staff Autonomy

In practice, the school’s organizational structure

was loosely coupled. An organization or parts of

an organization are loosely coupled when staff or

entities function independent of one another (Weick

1976). Staff in such systems have the freedom to

perform their duties with little oversight, and this

autonomy promotes creativity and job satisfaction,

but it can also inhibit communication (Eck and

Goodwin 2010; Firestone 1984; Pinelle and Gutwin

2006; Scheid-Cook 1990; Weick 1976).

Teacher autonomy. As in other high schools

across the country, teachers in this case had the

authority to handle student behavior problems in

their classrooms (Lortie 2002: Newman et al.

2004; Weick 1976). A foreign language teacher

described how she handled JD’s inappropriate

behavior prior to his attack:

[One day in class] we watched a little video

about the celebrations [for the] Day of the

Dead in Mexico. And [he] raised his hand,

and I thought he had a question. And then

he said, ‘‘So when can we start drinking

tequila in here?’’ And several students

said, ‘‘That is really disrespectful in Mrs.

[Smith’s] class.’’ And . . . he said ‘‘F—

you.’’ So, I walked over to his desk, and I

said, ‘‘[Y]ou may not talk like that in class.

I need that behavior to stop’’ . . . and then I

e-mailed his mom. (Teacher 1)

When asked if she relayed this incident to the

school’s principal or assistant principal, she said:

I did not [report it to the principal or assis-

tant principal] . . . [his behavior in my class]

wasn’t to the point where I needed to get an

administrator involved. . . . [T]he [prob-

lems] were things like I would say,

‘‘[A]re you listening?’’ Because he was

always talking [and] he was always on his

iPad or tablet. And I had to redirect [him]

on a pretty regular basis to get back in focus

with the class. At least once every class

period . . . which is not unusual with teen-

agers. . . . And he would say, ‘‘Yes,

ma’am,’’ in this really kind of arrogant

voice and he did that every day. (Teacher 1)

Although disrespectful and arrogant, these low-

level behavior problems did not warrant an admin-

istrator’s involvement (see Durkheim [1895]

1966; Vaughan 2016). This teacher had the

authority to handle this and other students’ disrup-

tive classroom behavior on her own and without

input from others (see also Duke, Showers, and

Imber 1980; Myers 1973). Hallett (2010) argues

that this autonomy grants teachers status as profes-

sionals and a reassuring sense of order.

A teacher’s ability to make these classroom

decisions in busy school settings is both practical

and efficient (see Weber 1978). For practical rea-

sons, teachers try to address behavior problems

quickly and reinforce classroom norms and values

consistently. For efficiency reasons, teachers often

need to handle these problems without consulting

an administrator because a teacher who regularly

involves administrators in student behavior prob-

lems may be viewed as ineffective, overwhelmed,

or unskilled by both students and administrators.

The school’s division of labor supported and the

principal reinforced this approach (see Weber

1978). The Principal explained:

Specifically, from [this teacher], no [she

didn’t need to report it]. She is very good

264 Sociology of Education 95(4)

with kids, and I think she has a good, strong

relationship with her students and handles

things in her classroom and is great about

contacting parents and things like that. So,

if she felt like it was resolved through con-

versations with the parents, and they were

notified, [I would] not necessarily [expect

to hear about it].

Thus, although not a formal school policy, the

school’s structure (e.g., division of labor, loose

coupling) created a culture (e.g., norms) that

encouraged teachers to handle low-level behavior

problems without consulting an administrator

(Becker 1953; McPherson 1972). These deeply

engrained cultural norms created a small but sig-

nificant gap in the school’s knowledge about the

student’s pattern of troubling behavior, and these

cultural norms represented the glue that connected

the school’s structure to staff’s daily work

activities.

The upside to localizing problems and solu-

tions is that it empowers staff. Using local control,

the system ‘‘can isolate its trouble spots and pre-

vent trouble from spreading’’ to other parts of

the school or system (Weick 1976:7). Localizing

problems and solutions for students of concern,

however, has two major downsides. The first

downside to localization is that each teacher has

to ‘‘negotiate separately’’ with the student to

address and redirect the same behavior problem

(Weick 1976:7). As a result, one teacher’s deci-

sion about a student’s disruptive classroom behav-

ior does not address the student’s recurring diffi-

culties, which may emerge in other classrooms.

In this case, JD’s difficulties included bullying

classmates in International Relations class; talking

about his genitals in a debate competition; show-

ing a teacher and peers a picture of his gun, which

he named the ‘‘Kurt Cobain’’; telling a teacher,

‘‘[I have] always been someone’s bitch . . . why

wouldn’t I make him my bitch after all that has

been done to me?’’; saying ‘‘teachers [are] out

to get me’’ and ‘‘my peers have often pushed

me’’; and telling a classmate to ‘‘go cut yourself.’’

To address the bullying behavior, JD’s Interna-

tional Relations teacher asked JD to stay after

class; JD did not stay, which struck the teacher

as ‘‘very unusual’’ (Teacher 2). Explaining that

students rarely disregarded his requests to stay after

class, he decided, ‘‘I needed to further investigate

who the student [was] . . . [to figure out] what

made him tick and . . . what his problem was . . .

[to know] how to reach [him]’’ (Teacher 2).1

The teacher talked with other teachers in his

department to figure out how to connect with the

student and manage his classroom conduct; he

developed a plan for JD. These independent nego-

tiations, however, did not address JD’s underlying

issues (e.g., anger) or recurring behavior problems

(e.g., threats, inappropriate comments; Weick

1976). This teacher resisted the ‘‘normalization

of deviance’’ in his classroom, but his efforts did

not come to the attention of the larger school orga-

nization (Vaughan 2016). When problems and sol-

utions remain local, ‘‘it [can] be difficult for the

loosely coupled system to repair the defective ele-

ment’’ (Weick 1976:7). In the end, each teacher’s

separate, independently brokered negotiation did

not solve the ‘‘real’’ problem underlying the stu-

dent’s disruptive behavior (e.g., anger, grievances).

The second downside to teachers’ independent

negotiations with troubled students is that school

administrators never learn about the full extent

of a student’s behavior problems or boundary test-

ing, making it difficult to see a pattern or escala-

tion. In everyday practice, organizations support

and even take for granted the routines that make

work efficient. Sociologists, however, note that

these routines can impede our ability to imagine

an alternative approach (Edelman et al. 2011;

Vaughan 2016). Thus, although educator auton-

omy is highly valued for classroom management,

it may limit the ability of other school staff to

learn about and effectively address a troubled stu-

dent’s warning signs. By the time JD yelled ‘‘I’m

going to kill that guy’’ (referring to the debate

coach who had demoted him) in his senior year

of high school, the intervention provided by the

school’s threat assessment team ended up being

‘‘too little [and] too late’’ (Weick 1976:7). These

examples shed light on the ways a school’s loosely

coupled organizational structure can foster a cul-

ture of autonomy for teachers, supporting indepen-

dent decision-making about students of concern.

These examples also reveal the way a student of

concern may fall through the cracks between an

organization’s structure and educators’ practices.

Administrator autonomy. Like teachers,

school administrators have considerable autonomy

(Eck and Goodwin 2010; Meyer and Rowan

1978). The principal and assistant principal in

this case held discretion over what behavior

Goodrum et al. 265

problems to investigate, whether and how to disci-

pline students, and whether to notify staff of stu-

dent behavior problems. Administrators’ auton-

omy originated in the school’s organizational

structure (e.g., procedures, task segregation, staff

hierarchy) and was reinforced through cultural

norms and beliefs. For example, the school’s

structure did not provide a clear or standardized

process for sharing a student’s threat assessment

results or reasons for suspension with teachers

(e.g., student database, email notifications). Dur-

ing a deposition, JD’s International Relations

teacher was asked, ‘‘[I]s that typical for you to

not know whether or not a student in your class

has been suspended?’’ He replied:

Well, usually what happens is I’ll get an

e-mail with a form I can print out, and it

will say, ‘‘This student has been sus-

pended.’’ It will not say why, but it will

say the length of time they’re suspended

and how [they will need] to make up their

assignments. (Teacher 2)

In a follow-up question, he was asked, ‘‘[D]o you

believe that it would be useful and helpful for the

teachers, such as yourself, to be told . . . why

[a student was] suspended?’’ The teacher replied:

You know, as teachers, I mean, we’re kind

of . . . on both sides of this issue. So, it’s

hard for me to really respond, I mean,

because we all have different jobs. And

so, as a teacher, I think sometimes we

want to know why, but I also know there’s,

you know, [FERPA] rights involved and

reasons why we can’t. (Teacher 2)

This segregation of tasks across ‘‘different jobs’’

shaped perceptions of roles and responsibilities.

In this case, task segregation fostered the view

that teachers did not have the right to know about

the problems leading to a student’s discipline even

when safety concerns arose (see Fox and Harding

2005; Weber 1978). In another study to come from

this case, administrators and teachers noted their

widespread confusion about the safety exception

within the FERPA, creating a code of silence

among staff (Goodrum et al. 2017).

The hierarchy of staff in the school also shaped

expectations and norms about information sharing.

The attacker’s foreign language teacher was asked

if she expected the assistant principal to ask her

whether to suspend JD for his outburst in her class.

She said, ‘‘I didn’t expect to be asked. . . . I think

that’s [the administrator’s] job . . . and I didn’t

have prior knowledge about [the student’s] behav-

ior’’ (Teacher 1).

An unintended consequence of the task segre-

gation and staff hierarchy in bureaucratic organi-

zations is that lower ranked staff may feel reluc-

tant to question a supervisor’s decision. In this

case, the organization’s policies became tightly

coupled with cultural norms. The debate coach

explained: ‘‘I wasn’t really in a position to agree

or disagree [with the assistant principal’s decision

to not suspend the student after he threated to kill

me]. All I could do was accept it.’’

The blanket acceptance of administrators’

decisions likely arises for two reasons. First,

teachers may grant administrators a range of

decision-making behaviors or ‘‘zone of accep-

tance’’ they are willing to accept (Kunz and Hoy

1976). The responsibility for student discipline in

the school fell to administrators, and the responsi-

bility for student management in classrooms fell

to teachers. Second, teachers may hesitate to ques-

tion an administrator’s decision due to expectations

of ‘‘reciprocal autonomy.’’ Teachers may grant

administrators autonomy over disciplinary actions

in exchange for their own autonomy over class-

rooms (Duke et al. 1980).

Granting a zone of acceptance and reciprocal

autonomy foster collegial relationships among

school staff, but they also create a culture where

staff cannot review or question each other’s deci-

sions. These powerful norms and beliefs about the

sanctity of educators’ autonomy elucidate the way

an organization’s loosely coupled structure gets

embedded in a school’s cultural fabric and daily

activities. Edelman and colleagues (2011:890)

note, ‘‘[C]ertain organizational structures become

widely institutionalized and taken for granted as

rational forms of organizational governance,’’

making it difficult for workers to question and

even imagine an alternative arrangement.

Structural Secrets about a Student of Concern

In loosely coupled systems, information fails to

reach the right people (Doyle 2010; Vaughan

2016). Fox and Harding (2005:73) note, ‘‘as

a whole, schools have access to a variety of infor-

mation about students—information about [their]

266 Sociology of Education 95(4)

physical health, academic progress, disciplinary

history, relationships with peers. . . . But since

tasks are highly segregated in schools, that infor-

mation is spread over multiple actors within the

school.’’ This task segregation creates uninten-

tional ‘‘secrets’’ within an organization’s struc-

ture. Fox and Harding (2005:85) continue, ‘‘we

prize privacy [in schools]—and for good

reason—but we do so occasionally at a high

cost.’’ The cost arises for two interrelated reasons.

First, connecting the dots of concern about a stu-

dent proves challenging when that concern is scat-

tered across people, classrooms, and time periods.

Second, knowing the context for a student’s cur-

rent behavior creates the opportunity for the staff

encountering the student to provide compassionate

understanding and effective intervention across

the system.

Connecting the dots. The culture of auton-

omy represents one way that gaps arise between

an organization’s structure and staff decisions,

and information silos represent another. Prior

research on school attacks notes the silos of infor-

mation that can emerge in school settings (Fox

and Harding 2005; Newman et al. 2004). These

findings highlight how those silos emerged and

how they hindered the assessment and manage-

ment of the student in this case. The foreign lan-

guage teacher explained:

Part of the problem . . . is that we’re very

compartmentalized. So, I might tell one of

my coworkers in the [foreign language]

department about [a student], but it

wouldn’t leave the [foreign language]

department. I wouldn’t walk over to math,

you know, for instance. I didn’t know about

[the math teacher’s] situation [with JD] at

all. (Teacher 1)

Thus, this teacher’s and others’ concerns about JD

remained in their respective departments. Despite

exhibiting a series of warning signs in the months

leading up to his deadly attack, no one understood

the seriousness or totality of those warning signs.

The debate coach, the attacker’s intended target,

explained:

I was working in isolation in the fall of

2013 when all of this was happening [with

JD]. It appears that [the foreign language

teacher] was having problems with [him]

and that [his junior year math teacher] had

had some previous problems [with him]

and [his senior year math teacher] had

some issues. None of us knew this. None

of the faculty that had direct interaction

with the student was aware of [his behavior

problems].

During the law enforcement investigation fol-

lowing the attack, school administrators learned

that campus security officers had witnessed the

student viewing photos of firearms on his laptop

in the cafeteria. Campus security officers reported

their observations to an assistant principal, but the

assistant principal never relayed that information

to the student’s threat assessment team or the

school’s principal. The principal acknowledged:

[An assistant principal] was told [that the

student was seen looking at photos of

guns online on his laptop in the school caf-

eteria in October] by a campus [security]

supervisor. . . . [After the shooting] I just

asked him why he never told anyone.

Why didn’t you let us know? Why didn’t

[the assistant principal who conducted the

threat assessment] know? And he said he

didn’t think it was that bad. . . . I think

[one assistant principal] knew some things

about [the student] and [the other assistant

principal] knew some things about [the stu-

dent but] never did it all come together.

The assistant principal who conducted the threat

assessment expressed frustration over not learning

about campus security’s observations: ‘‘I wish you

would have told me [that the student had been seen

viewing photos of guns on his laptop in the cafete-

ria] . . . you knew we had a threat assessment on

[the student] . . . I would have liked to have known

that.’’

Indeed, research on school shootings consis-

tently finds that attackers exhibited warning signs

among classmates, in classrooms, and at home, but

these warning signs fell through the cracks in the

systems for violence prevention (Goodrum et al.

2017; Langman and Straub 2019; MSDHSPSC

2019; Virginia Tech Review Panel 2007). The

postattack investigation in the Marjory Stoneman

Douglas case revealed that the attacker had 69

incidents of talking about weapons, engaging in

Goodrum et al. 267

concerning behavior, and threatening or engaging

in violence in local law enforcement’s records and

55 incidents of disciplinary referrals in the school

district’s records (MSDHSPSC 2019:234, 243).

Some might try to blame JD’s teachers for not

sounding the alarm sooner in this case, but this

response would neglect the fact that the school’s

organizational structure lacked the procedures

and cultural norms to support proactive communi-

cation about and interventions for students of con-

cern. The school resource officer in this case

lamented:

[I wish] we had more information given to

us about students . . . like an information

vortex . . . where everything [is] brought

together and where law enforcement is

involved, the therapist outside of the school

[is] involved. . . . [This] information needs

to be shared with everybody. Everybody

needs to be brought in.

The public health approach to school violence sup-

ports the idea of cross-agency communication for

addressing the difficult challenges facing students

and their families (Office of the Child Advocate

2014). To support this type of communication,

schools need policies (e.g., Memorandum of

Understanding), trusted partnerships (e.g., law

enforcement, social services), and mechanisms

(e.g., monthly meetings, emails) that can prove

difficult to broker in underresourced schools and

communities.

Knowing the context. The second reason

structural secrets hinder violence prevention in

schools relates to context. When critical informa-

tion about a student’s concerning behavior

remains unknown to teachers and other staff,

they may misinterpret observed behavior prob-

lems as isolated incidents, not as evidence of

a recurring or escalating problem. Context is crit-

ical to the way people perceive and process infor-

mation, particularly unusual information. When

we do not know the background or history of a stu-

dent’s concerning behavior (e.g., threats, anger,

aggression, weapons use), it becomes easy to nor-

malize, misinterpret, and dismiss moderately trou-

bling interactions and communications.

In violence prevention, having general knowl-

edge of the warning signs for violence proves

helpful for identifying students of concern (Fein

et al. 2004; NTAC 2019); having knowledge of

a specific student’s history of warning signs

proves critical to understanding and interpreting

that student’s subsequent behavior and communi-

cations. The student’s foreign language teacher

recalled, ‘‘No, absolutely not [the assistant princi-

pal did not tell me that JD had been the subject of

a threat assessment]. . . . No one in my department

knew’’ (Teacher 1).

Without knowing of JD’s threat assessment,

teachers and other staff downplayed his angry out-

bursts as concerning but not alarming. When

asked to explain the rationale for not informing

teachers about JD’s threat assessment results, the

principal said:

[Teachers] would know that a student was

suspended, though they may not know the

reason why . . . but we have explicitly said

multiple times in staff meetings [and] train-

ing that we’ve had with the staff around dis-

cipline, ‘‘if they ever have a question to

come see an administrator [or] a counselor.’’

But school staff did not realize they had questions

about the student’s behavior until after the shoot-

ing because they had no context with which to

interpret his problem behavior.

Without information on the big picture, observ-

ers can only see the smaller picture right in front of

them (see Costa 2012). The debate coach in the

case explained:

Not every student that screams and yells is

going to turn violent; so, we can’t . . . target

every kid that raises their voice. But we’ve

got to be able to talk among ourselves and

figure out—When is this a concern? When

is this justified?—to follow up about stu-

dents and concerns.

This discussion and follow-up can prove critical

when monitoring and supporting a threat-assessed

student. The debate coach continued:

I was fumbling in the dark [in dealing with

JD]. [And] I didn’t know what to do, and I

tried my best. I don’t think anybody [on the

school’s staff] did anything illegal. [But] I

think maybe some bad decisions

were made, bad judgment calls [were

made]. . . . I don’t know that we could

268 Sociology of Education 95(4)

have done anything to [stop him], but we

could have maybe tried harder.

Of course, ‘‘trying harder’’ remains admirable,

particularly for school staff juggling multiple

responsibilities, but it proves incredibly challeng-

ing within a system that lacks the organizational

structure and cultural norms to support those

efforts. Expecting or asking school staff to ‘‘try

harder’’ can never compensate for a system ill-

equipped to gather and share information about

a student of concern.

In a loosely coupled organization, there are

often reasonable explanations for the decision not

to share information (Vaughan 2016). Identifying

these reasonable explanations helps map the con-

nection between organizational structures and cul-

tural norms. The school’s principal explained:

[Behavioral and disciplinary problems were

not shared with teachers prior to the shoot-

ing because] it just had been the practice . . .

we wanted to make sure that kids weren’t

inappropriately judged by teachers, that

they were given a fair shot whether they

moved from, you know, one teacher to

another from year to year, that if something

happened in one classroom, it didn’t neces-

sarily mean that it was going to happen

again. [We wanted to] give them a fresh start

with a new teacher, that kind of thing, really

just erring more on the side of protecting the

kid, giving them a second chance.

The ‘‘fresh start mentality’’ means negative infor-

mation about the student gets lost within the

organization’s structure, and as a result, any future

negative information about the student is under-

valued and misinterpreted (Fox and Harding

2005). When school staff do not have all the rele-

vant information—whether it is mental health

concerns, behavioral concerns, or medical

history—they cannot effectively evaluate, inter-

vene, and monitor students. The challenge

remains, however, in what information should be

provided to whom when safety concerns arise.

Tightly coupling (or recoupling) the way infor-

mation about students of concern is shared through

formal policies and strict oversight may create

other problems (Hallett 2010). For example, shar-

ing information about a student’s threat assess-

ment results could prejudice a teacher against the

student so that benign behavior is misinterpreted

as malicious; at the same time, those results could

provide a teacher with insight that prompts a com-

passionate response and tailored intervention. The

manner in which this type of sensitive information

is treated (e.g., respectful vs. gossip) is shaped by

the culture (i.e., norms and values) and climate

(e.g., feelings) of the organization (Fein et al.

2004; Reason 1997; Vaughan 2016). Despite the

district’s tightly coupled threat assessment guide-

lines, the school’s loosely coupled structure

shaped cultural beliefs about what information

could be shared and with whom, hindering the pre-

vention of violence in this case.

CONCLUSION

Nearly 20 years ago, the U.S. Secret Service and

U.S. Department of Education developed guide-

lines for threat assessment and management for

students of concern, using findings from 37 school

attacks (Fein et al. 2004). These guidelines have

provided schools across the country with a frame-

work for how to evaluate and handle students of

concern (Fein et al. 2004; NTAC 2018), and sev-

eral national organizations have supplemented

these guidelines with best practices for threat

assessment in K–12 settings (NASP School Safety

and Crisis Response Committee 2021). The recent

increase in school attacks in the United States

(Irwin et al. 2021, 2022), however, raises ques-

tions about our approach to violence prevention

in U.S. schools. How is it possible that school vio-

lence could be on the rise despite federal guidance

and evidence-based practices for school safety?

One answer lies in the structure and culture of

the organizations where these tragic accidents

occur (Fox and Harding 2005; Newman et al.

2004). An ‘‘organizational accident’’ refers to

a ‘‘complex event [where] small mistakes com-

bine with each other and with latent conditions

hidden in the system to produce [an] unexpected

traged[y]’’ (Doyle 2010:145). In this study, the

‘‘small mistakes’’ related to the way the threat

assessment team evaluated and managed the trou-

bled student and the way teachers and administra-

tors handled the student’s concerning behavior in

isolation. Calling educators’ encounters ‘‘mis-

takes,’’ however, misrepresents the fact that the

school’s organizational structure remained ill-

equipped to support threat assessment. The latent

conditions in the school system included the

tightly coupled threat assessment guidelines, the

Goodrum et al. 269

loosely coupled organizational structure, and

the culture of educator autonomy.

On paper, tightly coupling the U.S. Secret

Service’s federal guidelines for threat assessment

to the school district’s policies on threat assess-

ment gave the appearance of compliance with

national recommendations for safety, creating an

institutional myth (i.e., symbolic and ceremonial

legitimacy; Hallett 2010; Meyer and Rowan

1977, 1978). In practice, the school’s loosely cou-

pled structure created gaps between threat assess-

ment guidelines and workers’ activities. The dis-

tance between policy and practice is common,

particularly in busy school settings where educa-

tors juggle competing concerns with limited

resources (Osher 2012). In their study of more

than 3,000 school-based prevention programs in

the United States, Gottfredson and Gottfredson

(2002) found that only 44 percent of evidenced-

based programs met a standard for effective imple-

mentation. Increasing the bureaucratic controls

over threat assessment in schools may sound like

the obvious solution to violence prevention in

this and other cases, but research suggests that

increasing control does not necessarily yield safety

improvements (Perrow 1984), particularly for

complex issues like violence prevention (Costa

2012). In fact, rigid guidelines may produce unin-

tended consequences, such as an increased use of

threat assessment with some student groups but

not others (e.g., along lines of race/ethnicity) or

frustration and turmoil among staff (see Hallett

2010; Porowski et al. 2014).

When reflecting on the lessons learned in this

case, the attacker’s foreign language teacher

said, ‘‘I think there should be a way that we

know of every student in trouble in that school.

. . . [I]t’s a big school and there’s a lot of kids.

And I think the more of us that are watching after

these sweet creatures . . . [the] better’’ (Teacher 2).

Watching after concerning students remains chal-

lenging when schools and districts do not have

the structural or cultural conditions to support

the institutionalization of evidence-based violence

prevention programs and procedures. How can

schools and communities balance the tightly cou-

pled procedures for threat assessment with person-

alized strategies for addressing students’ concern-

ing behaviors in underresourced school settings?

The U.S. Secret Service, U.S. Department of Edu-

cation, and others have argued that a positive

school culture and climate is key to school safety

(Goodrum et al. forthcoming; NTAC 2019;

Vossekuil et al. 2002). This emphasis on culture

and climate, however, neglects the reality that in

bureaucracies, the organizational structure

shapes the culture and climate. Thus, efforts to

prevent violence in schools should start by

improving the structure first and the culture and

climate second. A school’s organizational struc-

ture creates a strong force shaping the culture

and workers’ actions (see Weber 1978).

LIMITATIONS

A case study, like other methods of data collec-

tion, has its drawbacks (for a discussion, see Har-

ding et al. 2002; Roth and Mehta 2002). To mini-

mize the effect of these limitations on the project,

we followed Roth and Mehta’s (2002) suggestions

by gathering data from multiple sources (e.g., law

enforcement interviews, depositions, school

records) and by investigating conflicting reports

through an analysis of school records, deposition

testimony, and the law enforcement investigation.

Also, during data analysis, facts were confirmed

through a triangulation of data using multiple

deponents and official documents.

RECOMMENDATIONS

Sociobiologist Rebecca Costa (2012) has argued

that difficult and persistent social problems, like

school violence, require a comprehensive strategy

whereby multiple solutions are implemented simul-

taneously within social institutions and the culture.

The findings from this study suggest three recom-

mendations for future safety efforts in schools.

Fortify the Structure

First, when existing solutions do not resolve the

problem, conduct a gap analysis to consider

whether the organizational structure has the capac-

ity (e.g., resources, motivation) to effectively

implement those solutions. The field of implemen-

tation science consistently finds that delivering

evidence-based programs proves challenging in

busy school settings, where staff remain underre-

sourced and overworked (Greenberg 2010;

Mihalic and Irwin 2003; Osher 2012). At the

school level, a school’s capacity to implement

a new program can be measured and improved,

which would likely benefit implementation and

outcomes (Kingston et al. 2018).

270 Sociology of Education 95(4)

At the societal level, research should seek to

identify the common structural weaknesses in vio-

lence prevention efforts across U.S. schools. How

many other schools have gaps between guidelines

and practices in threat assessment and other vio-

lence prevention programs? Using this evidence,

state and federal governments could explore

ways to provide schools with the financial support

and implementation assistance needed to help

institutionalize best practices for violence preven-

tion and threat assessment (Greenberg 2010).

Finally, schools could use existing institutional-

ized procedures in the academic realm, such as

the steps for building an Individualized Education

Program (IEP) for students with learning needs, as

a model for how to approach students with behav-

ioral and mental health needs. An IEP-like plan for

students with behavioral and mental health needs

would provide educators with information about

and guidance on how to best support these stu-

dents in a consistent and compassionate manner,

which would yield improved academic outcomes.

Leverage the Culture

Second, schools could consider leveraging the

deeply valued culture of autonomy to empower

educators to recognize and report the warning

signs for violence, including threats, intense anger,

weapons interest, bullying victimization, depres-

sion, and harassing others (NTAC 2019). Educator

training on the warning signs for violence could

include (a) an explanation of the difference

between disrespectful adolescent behavior and

the warning signs for violence, (b) the need to

report warning signs to school administrators, (c)

the purpose for and role of teachers in threat

assessment, and (d) guidance on how to prevent

discrimination and bias in reporting warning signs

(NTAC 2019; Reeves and McCarthy 2021).

Share the Responsibility

Finally, as a society, we need to recognize that the

prevention of violence is a social problem, not

a school problem, and this social problem requires

a comprehensive approach with multicomponent

strategies (e.g., reduce access to firearms, educate

bystanders on anonymous reporting, implement

programs for violence prevention). To share the

responsibility for addressing the warning signs

for violence, communities should consider devel-

oping community-based threat assessment teams

that work with school-based threat assessment

teams to identify and manage individuals of con-

cern. The public health approach to violence pre-

vention supports a multidisciplinary perspective

where community stakeholders (e.g., educators,

law enforcement officials, social services profes-

sionals, and criminal justice officials) come

together to define the problem, consider the fac-

tors influencing the problem, and create and

implement strategies to address the problem (Cen-

ters for Disease Control and Prevention 2022). We

can no longer expect school administrators, teach-

ers, and school psychologists to address the prob-

lem of violence prevention on their own. The com-

plexity of the school violence problem, the

financial implications of the solutions, and the

fear of an innovative comprehensive approach

must not paralyze us (Costa 2012). We have the

knowledge to build organizational structures and

cultures of safety; it is time to translate that knowl-

edge into institutionalized everyday practice.

ACKNOWLEDGMENTS

The authors are grateful to the parents of the victim in

this case, who sought an investigative arbitration with

the school district to learn lessons on how to prevent

the type of violence that took their daughter. We are

also indebted to the school and district officials who

gave testimony, helping us to reflect on and improve vio-

lence prevention practices in schools.

FUNDING

The study was funded by a grant from the Denver

Foundation.

RESEARCH ETHICS

Institutional review board approval was not required

because the authors did not recruit or interview subjects.

In addition, all parties in the case agreed to the public

release of the project data to share lessons learned on

violence prevention.

ORCID iDs

Sarah Goodrum https://orcid.org/0000-0002-9529-5563

Jessie Slepicka https://orcid.org/0000-0002-9328-5716

NOTE

1. The sheriff’s report noted, ‘‘[The teacher] wanted to

know how to deal with [the student] because [he] was

telling other kids they were stupid. [He] was

Goodrum et al. 271

concerned because [the student] is . . . being seen as

a ‘verbal bully’’’ (Exhibit 14).

REFERENCES

Alper, Mariel, and Lauren Glaze. 2019. Source and Use

of Firearms Involved in Crimes: Survey of Prison

Inmates, 2016. Washington, DC: U.S. Department

of Justice, Office of Justice Programs, Bureau of Jus-

tice Statistics. https://bjs.ojp.gov/content/pub/pdf/

suficspi16.pdf.

Becker, Herbert. 1953. ‘‘The Teacher in the Authority

System of the Public School.’’ The Journal of Edu-

cational Sociology 27(3):128–41.

Bidwell, Charles E. 1965. ‘‘The School as Formal Orga-

nization.’’ Pp. 972–1018 in Handbook of Organiza-

tions, edited by J. G. March. Chicago, IL: Rand

McNally.

Biglan, Anthony, Patricia Mrazek, Douglas Carnine, and

Brian R. Flay. 2003. ‘‘The Integration of Research

and Practice in the Prevention of Youth Problem

Behaviors.’’ American Psychologist 58(6–7):

433–40. doi:10.1037/0003-066X.58.6-7.433.

Birkinshaw, Julian, Mary Yoko Brannen, and Rosalie L.

Tung. 2011. ‘‘From a Distance and Generalizable to

up Close and Grounded: Reclaiming a Place for

Qualitative Methods in International Business

Research.’’ Journal of International Business Studies

42:573–81.

Blair, J. Pete, and Katherine W. Schweit. 2014. A Study

of Active Shooter Incidents, 2000-2013. Washington,

DC: Texas State University and Federal Burau of

Investigation, U.S. Department of Justice.

Bradley, Elizabeth H., Leslie A. Curry, and Kelly J.

Devers. 2007. ‘‘Qualitative Data Analysis for Health

Services Research: Developing Taxonomy, Themes,

and Theory.’’ Health Services Research 42(4):

1758–72.

Bryman, Alan, and Emily Bell. 2011. Business Research

Methods. 3rd ed. Oxford, UK: Oxford University

Press.

Burnard, Philip, Paul Gill, Kate Stewart, Elizabeth Trea-

sure, and Barbara Chadwick. 2008. ‘‘Analyzing and

Presenting Qualitative Data.’’ British Dental Journal

204(8):429–32.

Burrows, Dustin, Joe Moody, and Eva Guzman. 2022.

Investigative Committee on the Robb Elementary

Shooting, Interim Report 2022. Austin, TX: House

of Representatives. https://static.texastribune.org/

media/files/d005cf551ad52eea13d8753ede93320c/Uv

alde%20Robb%20Shooting%20Report%20-%20Tex

as%20House%20Committee.pdf?_ga=2.253815820

.5600916.1658677320-1012249026.1658677320.

Centers for Disease Control and Prevention. 2022. The

Public Health Approach to Violence Prevention.

Atlanta, GA: Centers for Disease Control and

Prevention. https://www.cdc.gov/violencepreven

tion/pdf/PH_App_Violence-a.pdf.

Chapman, Katrina. 2009. ‘‘A Preventable Tragedy at

Virginia Tech: Why Confusion Over FERPA’s Pro-

visions Prevents Schools from Addressing Student

Violence.’’ Public Interest Law Journal 18:349–85.

Condon, Bernard. 2022. ‘‘Texas School Shooter Left

Trail of Ominous Warning Signs.’’ Washington

Post, July 19. https://www.washingtonpost.com/nat

ional/texas-killer-earned-ominous-nickname-school-

shooter/2022/07/18/24962b94-06e5-11ed-80b6-43f

2bfcc6662_story.htmlhtmlfile/Shell/Open/Command.

Costa, Rebecca. 2012. The Watchman’s Rattle: A Radi-

cal New Theory of Collapse. Philadelphia, PA: Van-

guard Press.

Crowe, Sarah, Kathrin Cresswell, Ann Robertson, Guro

Huby, Anthony Avery, and Aziz Sheikh. 2011.

‘‘The Case Study Approach.’’ BMC Medical

Research Methodology 11(1):100–108.

Dörner, Dietrich. 1996. The Logic of Failure: Recogniz-

ing and Avoiding Error in Complex Situations. Cam-

bridge, MA: Perseus Press.

Doyle, James. 2010. ‘‘Learning from Error in American

Criminal Justice.’’ The Journal of Criminal Law and

Criminology 100(1):109–47.

Duke, Daniel L., Beverly K. Showers, and Michael

Imber. 1980. ‘‘Teachers and Shared Decision Mak-

ing: The Costs and Benefits of Involvement.’’ Edu-

cational Administration Quarterly 16(1):93–106.

Durkheim, Emile. [1895] 1966. The Rules of Sociologi-

cal Method. New York, NY: Free Press.

Eck, James, and Bryan Goodwin. 2010. ‘‘Autonomy for

School Leaders.’’ School Administrator 67(1):24–27.

Edelman, Lauren B., Linda H. Krieger, Scott R. Eliason,

Catherin R. Albiston, and Virginia Mellema. 2011.

‘‘When Organizations Rule: Judicial Deference to

Institutionalized Employment Structures.’’ American

Journal of Sociology 117(3):888–954.

Erickson, William. 2001. The Report of Governor Bill

Owens’ Columbine Review Commission. Denver,

CO: The State of Colorado.

Federal Bureau of Investigation. 2016. Active Shooter

Incidents in the United States in 2014 and 2015.

Washington, DC: U.S. Department of Justice.

Federal Bureau of Investigation. 2018. Active Shooter

Incidents in the United States in 2018. Washington,

DC: The Advanced Law Enforcement Rapid

Response Training Center at Texas State University

and U.S. Department of Justice.

Federal Bureau of Investigation. 2019. Active Shooter

Incidents in the United States in 2019. Washington,

DC: The Advanced Law Enforcement Rapid

Response Training Center at Texas State University

and U.S. Department of Justice.

Fein, Robert, Bryan Vossekuil, William Pollack, Randy

Borum, William Modzeleski, and Marisa Reddy.

2004. Threat Assessment in Schools: A Guide to

272 Sociology of Education 95(4)

Managing Threatening Situations and to Creating

Safe School Climates. Washington, DC: United

States Secret Service and United States Department

of Education.

Firestone, William A. 1984. ‘‘The Study of Loose Cou-

pling: Problems, Progress, and Prospects.’’ Pp. 3–30

in Research in Sociology of Education and Socializa-

tion. Vol. 5, edited A. Kerckhoff. Greenwich, CT:

JAI Press.

Fixsen, Dean L., Sandra F. Naoom, Karen A. Blasé, and

Frances Wallace. 2005. Implementation Research: A

Synthesis of the Literature. Tampa: University of

South Florida, National Implementation Research

Network.

Flyvbjerg, Bent. 2006. ‘‘Five Misunderstandings about

Case-Study Research.’’ Qualitative Inquiry 12(3):

219–45.

Forman, Susan G., S. Serene Olin, Kimberly Eaton

Hoagwood, Maura Crowe, and Nao Saka. 2009.

‘‘Evidence-Based Interventions in Schools: Devel-

opers’ Views of Implementation Barriers and Facili-

tators.’’ School Mental Health 1:26–36. doi:10.1007/

s12310-008-9002-5.

Fox, Cybelle, and David J. Harding. 2005. ‘‘School

Shootings as Organizational Deviance.’’ Sociology

of Education 78(1):69–97.

Geis, Gilbert. 1991. ‘‘The Case Study Method in Socio-

logical Criminology.’’ Pp. 200–23 in A Case for the

Case Study, edited by J. R. Feagin, A. Orum, and G.

Sjoberg. Chapel Hill: University of North Carolina

Press.

George, Alexander L., and Timothy J. McKeown. 1985.

‘‘Case Studies and Theories of Organizational Deci-

sion Making.’’ Advances in Information Processing

in Organizations 2(1):21–58.

Gephart, Jr., and Robert P. 2004. ‘‘Qualitative Research

and The Academy of Management Journal.’’ The

Academy of Management Journal 47(4):454–62.

Gill, Peter Edward, and Max Allen Stenlund. 2006.

‘‘Dealing with a Schoolyard Bully: A Case Study.’’

Journal of School Violence 4:47–62.

Goodrum, Sarah, Mary K. Evans, Andrew J. Thompson,

and William Woodward. 2019. ‘‘Learning from

a Failure in Threat Assessment: Eleven Questions

and Not Enough Answers.’’ Behavioral Sciences &

the Law 37(4):353–71. doi:10.10002/bsl.2399.

Goodrum, Sarah, Beverly Kingston, Sabrina Arredondo

Mattson, Susanne Argamaso, Jody Witt, and

Amanda Matthews. Forthcoming. Balancing the

Components of a Comprehensive School Safety

Framework: Findings from NIJ’s Comprehensive

Safe School Initiative. Washington, DC: U.S. Depart-

ment of Justice, National Institute of Justice.

Goodrum, Sarah, Andrew J. Thompson, Kyle C. Ward,

and William Woodward. 2018. ‘‘A Case Study on

Threat Assessment: Learning Critical Lessons to

Prevent School Violence.’’ Journal of Threat

Assessment and Management 5(3):121–36. doi:

10.1037/tam0000104.

Goodrum, Sarah, William Woodward, and Andrew J.

Thompson. 2017. ‘‘Sharing Information to Promote

a Culture of Safety.’’ NASSP Bulletin 101(3):

215–40. doi:10.1177/0192636517727347.

Gottfredson, Denise C., and Gary D. Gottfredson. 2002.

‘‘Quality of School-Based Prevention Programs:

Results from a National Survey. Journal of Research

in Crime and Delinquency 39:3–35. doi:10.1177/

002242780203900101.

Greenberg, Mark T. 2010. ‘‘School-Based Prevention:

Current Status and Future Challenges.’’ Effective

Education 2(1):27–52.

Hallett, Tim. 2010. ‘‘The Myth Incarnate: Recoupling

Processes, Turmoil, and Inhabited Institutions in an

Urban Elementary School.’’ American Sociological

Review 75(1):52–74.

Hamm, Kevin. 2021. ‘‘Tracking Colorado’s Mass Shoot-

ings: 52 Have Died in Nine Incidents Since 1993.’’

The Denver Post, March 24. https://www.denver

post.com/2021/03/24/colorado-mass-shootings-inci

dents-list/.

Harding, David J., Cybelle Fox, and Jal D. Mehta. 2002.

‘‘Studying Rare Events through Qualitative Case

Studies: Lessons from a Study of Rampage School

Shootings.’’ Sociological Methods & Research

31(2):174–217.

Irwin, Veronique, Ke Wang, Jiashan Cui, and Alexandra

Thompson. 2022. Report on Indicators of School

Crime and Safety: 2021 (NCES 2022-092/NCJ

304625). Washington, DC: National Center for Edu-

cation Statistics, U.S. Department of Education, and

Bureau of Justice Statistics, Office of Justice Pro-

grams, U.S. Department of Justice. https://nces.ed

.gov/pubsearch/pubsinfo.asp?pubid=2022092.

Irwin, Veronique, Ke Wang, Jiashan Cui, Jizhi Zhang,

and Alexandra Thompson. 2021. Report on Indicators

of School Crime and Safety: 2020 (NCES 2021-092/

NCJ 300772). Washington, DC: National Center for

Education Statistics, U.S. Department of Education,

and Bureau of Justice Statistics, Office of Justice Pro-

grams, U.S. Department of Justice. https://nces.ed.

gov/pubsearch/pubsinfo.asp?pubid=2021092.

Keen, Justin, and T. Packwood. 1995. ‘‘Qualitative

Research: Case Study Evaluation.’’ BMJ 311:

444–46.

Kingston, Beverly, Sabrina Arredondo Mattson, Allison

Dymnicki, Elizabeth Spier, Monica Fitzgerald, Kim-

berly Shipman, Sarah Goodrum, William Wood-

ward, Jody Witt, Karl G. Hill, and Delbert Elliott.

2018. ‘‘Building Schools’ Readiness to Implement

a Comprehensive Approach to School Safety.’’ Clin-

ical Child and Family Psychology Review 21:

433–49. https://doi.org/10.1007/s10567-018-0264-7.

Kunz, Daniel W., and Wayne K. Hoy. 1976. ‘‘Leader-

ship Style of Principals and the Professional Zone

Goodrum et al. 273

of Acceptance of Teachers.’’ Educational Adminis-

tration Quarterly 12(3):49–64.

Langman, Peter, and Frank Straub. 2019. A Comparison

of Averted and Completed School Attacks from the

Police Foundation’s Averted School Violence Data-

base. Washington, DC: Office of Community Ori-

ented Policing Services.

Lortie, Dan C. 2002. Schoolteacher: A Sociological

Study. 2nd ed. Chicago, IL: University of Chicago

Press.

Mahoney, James. 1999. ‘‘Nominal, Ordinal, and Narra-

tive Appraisal in Macrocausal Analysis.’’ American

Journal of Sociology 104(4):1154–96.

Mahoney, James. 2000. ‘‘Strategies of Causal Inference

in Small N Analysis.’’ Sociological Methods and

Research 28(4):387–424.

Marjory Stoneman Douglas High School Public Safety

Commission. 2019. Initial Report. Tallahassee, FL:

Governor’s Office.

Maxwell, Joseph A. 2013. Qualitative Research Design:

An Interactive Approach. Thousand Oaks, CA: Sage

Publications.

McCutcheon, David, and Jack Meredith. 1993. ‘‘Con-

ducting Case Study Research in Operations Manage-

ment.’’ Journal of Operations Management 11:

239–56.

McPherson, Gertrude H. 1972. Small Town Teacher.

Cambridge, MA: Harvard University Press.

Meyer, John W., and Brian Rowan. 1977. ‘‘Institutional

Organizations: Formal Structure as Myth and Cere-

mony.’’ American Journal of Sociology 83(2):

340–63.

Meyer, John W., and Brian Rowan. 1978. ‘‘The Struc-

ture of Educational Organizations: Formal Structure

as Myth and Ceremony.’’ Pp. 78–109 in Environ-

ments and Organizations. Vol. 80, edited by M.

Meyer. San Francisco, CA: Jossy Bass.

Mihalic, Sharon F., and Katherine Irwin. 2003. ‘‘Blue-

prints for Violence Prevention: From Research to

Real-World Settings – Factors Influencing the Suc-

cessful Replication of Model Programs.’’ Youth Vio-

lence and Juvenile Justice 1(4):307–29.

Mills, Albert J., Gabrielle Durepos, and Elden Wiebe,

eds. 2010. Encyclopedia of Case Study Research, Vol-

umes I and II. Thousand Oaks, CA: Sage Publications.

Myers, Donald A. 1973. Teacher Power. Lexington,

MA: Lexington Books.

NASP School Safety and Crisis Response Committee.

2021. Behavioral Threat Assessment and Management:

Best Practice Considerations for K-12 Schools.

National Association of School Psychologists. https://

www.nasponline.org/btam

National Institute of Justice. 2020. A Comprehensive

School Safety Framework: Report to the Committees

on Appropriations. Washington, DC: U.S. Depart-

ment of Justice, Office of Justice Programs, National

Institute of Justice. https://www.ojp.gov/pdffiles1/

nij/255078.pdf.

National Threat Assessment Center. 2018. Enhancing

School Safety Using a Threat Assessment Model:

An Operational Guide for Preventing Targeted

School Violence. Washington, DC: U.S. Secret Ser-

vice, Department of Homeland Security.

National Threat Assessment Center. 2019. Protecting

America’s Schools: A U.S. Secret Service Analysis

of Targeted School Violence. Washington, DC: U.S.

Secret Service, Department of Homeland Security.

Newman, Katherine S., Cybelle Fox, Wendy Roth, Jal

Mehta, and David Harding. 2004. Rampage: The

Social Roots of School Shootings. New York, NY:

Perseus Books.

Office of the Child Advocate. 2014. Shooting at Sandy

Hook Elementary School: Report of the Office of the

Child Advocate. Hartford, CT: State of Connecticut.

Osher, David. 2012. ‘‘Commentary: Implementation in

Busy Kitchens and Swampy Lowlands.’’ Social Pol-

icy Report 26(4):23–24.

Paino, Maria. 2018. ‘‘From Policies to Principals: Tiered

Influences on School-Level Coupling.’’ Social

Forces 96(3):1119–54.

Perrow, Charles. 1984. Normal Accidents: Living with

High-Risk Technologies. New York, NY: Basic Books.

Pinelle, David, and Carl Gutwin. 2006. ‘‘Loose Coupling

and Healthcare Organizations: Deployment Strate-

gies for Groupware.’’ Computer Supported Coopera-

tive Work 15(5–6):537–72.

Planty, Michael, Duren Banks, Christine Lindquist, Joel

Cartwright, and Amanda Witwer. 2020. Tip Lines for

School Safety: A National Portrait of Tip Line Use.

Research Triangle Park, NC: RTI International.

Porowski, Allan, Rosemarie O’Conner, and Aikaterini

Passa. 2014. Disproportionality in School Discipline:

An Assessment of Trends in Maryland, 2009–12

(REL 2014–017). Washington, DC: U.S. Department

of Education, Institute of Education Sciences,

National Center for Education Evaluation and

Regional Assistance, Regional Educational Labora-

tory Mid-Atlantic. http://ies.ed.gov/ncee/edlabs.

Randazzo, Marisa, Randy Borum, Bryan Vossekuil,

Robert Fein, William Modzeleski, and William Pol-

lack. 2006. ‘‘Threat Assessment in Schools: Empirical

Support and Comparison with Other Approaches.’’

Pp. 147–56 in Handbook of School Violence and

School Safety: From Research to Practice, edited by

S. Jimerson and M. Furlong. Mahwah, NJ: Lawrence

Erlbaum.

Reason, James. 1997. Managing the Risks of Organiza-

tional Accidents. Farnham, UK: Ashgate Publishing.

Reason, James. 1998. ‘‘Achieving a Safe Culture: The-

ory and Practice.’’ Work & Stress 12(3):293–306.

doi:10.1080/02678379808256868.

Reeves, Melissa, and Stephen Brock. 2018. ‘‘School

Behavioral Threat Assessment and Management.’’

Contemporary School Psychology 22(2):148–62.

Reeves, Melissa A., and Courtenay McCarthy. 2021.

Upholding Student Civil Rights and Preventing

274 Sociology of Education 95(4)

Disproportionality in Behavioral Threat Assessment

and Management. Bethesda, MD: National Associa-

tion of School Psychologists.

Ringwalt, Chris, Amy A. Vincus, Sean Hanley, Susan T.

Ennett, J. Michael Bowling, and Louise Ann Rohr-

bach. 2009. ‘‘The Prevalence of Evidence-Based

Drug Use Prevention Curricula in U.S. Middle

Schools in 2005.’’ Prevention Science 10:33–40.

Roth, Wendy D., and Jal D. Mehta. 2002. ‘‘The Rasho-

mon Effect: Combining Positivist and Interpretivist

Approaches in the Analysis of Contested Events.’’

Sociological Methods & Research 31:131–73. doi:

10.1177/0049124102031002002.

Scheid-Cook, Teresa. 1990. ‘‘Ritual Conformity and

Organizational Control: Loose Coupling or Profes-

sionalization?’’ The Journal of Applied Behavioral

Science 26(2):183–99.

‘‘School Shootings This Year: How Many and Where.’’

2022. Education Week, January 5. https://www

.edweek.org/leadership/school-shootings-this-year-

how-many-and-where/2022/01.

Stake, Robert E. 1995. The Art of Case Study Research.

London: Sage Publications.

Stake, Robert E. 2005. ‘‘Qualitative Case Studies.’’ Pp.

443–66 in Handbook of Qualitative Research. 3rd

ed., edited by N. Denzin and Y. Lincoln. Thousand

Oaks, CA: Sage Publications.

Strauss, Anslem. 1987. Qualitative Analysis for Social

Scientists. New York, NY: Cambridge University

Press.

Turner, Barry A. 1976. ‘‘The Organizational and Interor-

ganizational Development of Disasters.’’ Adminis-

trative Science Quarterly 21:378–97.

Thrush, Glenn. 2021. ‘‘‘Ghost Guns’: Firearm Kits

Bought Online Fuel Epidemic of Violence.’’ The

New York Times, November 20. https://www

.nytimes.com/2021/11/14/us/ghost-guns-homemade-

firearms.html.

U.S. Department of Education. 2011. Prevalence and

Implementation Fidelity of Research-Based Preven-

tion Programs in Public Schools: Final Report.

Rockville, MD: U.S. Department of Education.

Vaughan, Diane. 2016. The Challenger Launch Deci-

sion: Risky Technology, Culture, and Deviance at

NASA. Enlarged 2nd ed. Chicago, IL: University of

Chicago Press.

Virginia Tech Review Panel. 2007. ‘‘Mass Shootings at

Virginia Tech: Report of the Review Panel.’’

scholar.lib.vt.edu/prevail/docs/VTReviewPanelRe

port.pdf.

Vossekuil, Bryan, Robert Fein, Marisa Reddy, Robert

Borum, and William Modzeleski. 2002. The Final

Report and Findings of the Safe School Initiative:

Implications for the Prevention of School Attacks in

the U.S. Washington, DC: U.S. Secret Service and

Department of Education.

Weber, Max. 1978. Economy and Society. Berkley and

Los Angeles: University of California Press.

Weick, Karl. 1976. ‘‘Educational Organizations as

Loosely Coupled Systems.’’ Administrative Science

Quarterly 21(1):1–19.

Wintemute, Garen J. 2015. ‘‘The Epidemiology of Fire-

arm Violence in the Twenty-First Century United

States.’’ Annual Review of Public Health 36:5–19.

Yin, Robert. 2014. Case Study Research Design and

Methods. 5th ed. Thousand Oaks, CA: Sage

Publications.

Author Biographies

Sarah Goodrum, PhD, is a senior research associate at

the Center for the Study and Prevention of Violence at

the University of Colorado Boulder. Her main fields of

study include intimate partner violence, homicide vic-

timization, and violence prevention. She is currently

studying the relationship between organizational culture

and violence prevention and the impact of training on

threat assessment procedures.

Jessie Slepicka is a doctoral candidate in the Depart-

ment of Sociology and Criminology at The Pennsylvania

State University. His research interests include crimino-

logical theory, green criminology, comparative criminol-

ogy, spatial criminology, and quantitative research

methods.

William Woodward, MPA, is a research associate at the

University of Colorado’s Center for the Study and Pre-

vention of Violence and is former Colorado State Direc-

tor of Criminal Justice. He coauthored the Arapahoe

High School shooting investigation study and the Colo-

rado Attorney General’s School Safety Guide. He pre-

sented a TEDx talk on preventing school shootings in

2016.

Beverly Kingston, PhD, is director and senior research

associate at the Center for the Study and Prevention of

Violence at the University of Colorado Boulder. Her

main fields of interest are comprehensive public health

approaches to violence prevention, school safety,

neighborhood social processes, and implementation

science. She currently leads several school and commu-

nity initiatives and research studies focused on these

topics.

Goodrum et al. 275