ArticlePDF Available

Abstract and Figures

Active shooting violence at confined settings, such as educational institutions, poses serious security concerns to public safety. In studying the effects of active shooter scenarios, the common denominator associated with all events, regardless of reason/intent for shooter motives, or type of weapons used, was the location chosen and time expended between the beginning of the event and its culmination. This in turn directly correlates to number of casualties incurred in any given event. The longer the event protracts, the more casualties are incurred until law enforcement or another barrier can react and culminate the situation. Using AnyLogic technology, devise modeling scenarios to test multiple hypotheses against free-agent modeling simulation to determine the best method to reduce casualties associated with active shooter scenarios. Test four possible scenarios of responding to active shooter in a public school setting using agent-based computer modeling techniques-scenario 1: basic scenario where no access control or any type of security is used within the school; scenario 2, scenario assumes that concealed carry individual(s) (5-10 percent of the work force) are present in the school; scenario 3, scenario assumes that the school has assigned resource officer; scenario 4, scenario assumes that the school has assigned resource officer and concealed carry individual(s) (5-10 percent) present in the school. Statistical data from modeling scenarios indicating which tested hypothesis resulted in fewer casualties and quicker culmination of event. The use of AnyLogic proved the initial hypothesis that a decrease on response time to an active shooter scenario directly reduced victim casualties. Modeling tests show statistically significant fewer casualties in scenarios where on scene armed responders such as resource officers and concealed carry personnel were present.
Content may be subject to copyright.
1
Mitigating Active Shooter Impact;
Analysis for Policy Options Based
on Agent/Computer Based
Modeling
By: Charles Anklam, Adam Kirby, Filipo Sharevski and Dr. J. Eric Dietz
March 2014
2
Project Description
Active shooting violence at educational institutions is a phenomenon that poses serious security
concerns about public safety due to the horrifying outcome and potentially large number of causalities
and injured individuals stemming from such an event. US Department of Homeland Security has
described the active shooter as an:
..Individual actively engaged in killing or attempting to kill people in a confined and
populated area; in most cases, active shooters use firearms(s) and there is no pattern or
method to their selection of victims. 1
In relation to school settings, active shooter incidents typically take place in densely populated
areas within the school perimeter, such as a classroom, administration offices, or common areas like
cafeterias, gymnasiums and libraries. These incidents are unpredictable, evolve quickly, and have a main
goal of mass murdering, rather than other criminal conduct, such as robbery. In many cases, the
perpetrator is equipped with multiple weapons and tries to accomplish his goal in the minimum amount of
time. The shooter also typically does not have an escape plan, so he either commits suicide, surrenders,
or is engaged by law enforcement or other responding individual1. As real life evidence to active shooting
phenomena, Table 1 summarizes five incidents of active shootings at educational institutions that took
place in the last five years2, 3, 4, 5, 6.
In studying the effects of active shooter scenarios, the baseline for establishing a hypothesis is
the analysis of empirical data from previous active shooter incidents. The common denominator
associated with all events, regardless of reason or intent for shooter motives, or type of weapons used,
was the location chosen and time expended between the beginning of the event and its culmination2, 3, 4, 5,
6. This in turn includes and directly correlates to the number of casualties incurred in any given event.
The longer the event protracts, the more casualties are incurred until law enforcement or another barrier
can react and culminate the situation.
Given the fact that active shooting incidents can have severe consequences to public safety and
can result in significant causalities and injured individuals2, 3, 4, 5, 6, this research project employed the use
of computer based modeling to model and analyze four possible scenarios to address an active shooter
in a public school setting to determine which scenario reduces the most casualties:
Scenario 1 This is a basic scenario where no access control or any type of security is employed
within the school
Scenario 2 This scenario assumes that concealed carry individual(s) (5-10% of the work force)
are present in the school
Scenario 3 This scenario assumes that the school has an assigned Resource Officer
Scenario 4 This scenario assumes that the school has an assigned Resource Officer and that
there are concealed carry individual(s) (5-10% of work force) present in the school
The research methodology employs four varying scenarios that evaluate implemented barriers,
observing their effectiveness on an active shooter event reaching a culminating point. These barriers,
therefore, are directly correlated to the time (span of time) for which an event is allowed to exist before
being diffused. The intervening time therefore correlates to the number of casualties expected to be
3
inflicted during the time of the event. Using four different examples, the model allows the injection of
varying modes of blocks or barriers which can ultimately result in an event either ending sooner, or lasting
longer until final resolution is accomplished. This process then answers the hypothesis: “Does a
relationship exists between the number and types barriers injected into an active shooter scenario and
numbers of casualties incurred?”
As a main analysis method, agent based simulation models are developed in order to assess the
effectiveness of the employed security measures expressed with the number of causalities and injured
individuals, and response time of the first responders (time to arrive on scene and time to engage with the
shooter). These measurements of effectiveness were chosen since the historical data (including Table 1)
indicate that time is the most compelling factor in determining casualty rates for active shooter events.
Agent based modeling is chosen because it is the most suitable approach for accurate representation and
tracking the actions of the entities involved in the active shooting incident, primarily the shooter,
concealed carry individual(s), or the resource officers.
Further analysis of the proximity of the local police station to the modeled school, assumptions
about the weapons used by an active shooter, and the movement pattern of the shooter within the school,
allowed for identifying the possible security measures that could be employed in order to minimize the
number of causalities during an active shooting incident. Another purpose of this analysis is to evaluate
the model's ability to differentiate impacts between shelter-in-place and building evacuation during this
type of incident.
(Insert Table 1)
Background
The specific nature of the active shooting incident requires reconsideration of security and school
safety measures and polices. In this direction, there are several good practices1, 7 that can be employed
for coping with an active shooter situation. Department of Homeland Security recommendations relative
to the active shooter response1 include guidelines on how to respond when an active shooter is in the
school perimeter (identifying evacuation, hiding, or active engagement actions with the shooter), training
and preparing school staff for an active shooter situation (Emergency Action Plan and training exercises),
recognizing potential workplace violence and managing the consequences of an active shooter situation.
These recommendations are further impressed by MSA Security, an industry leader in security consulting
and management, who suggest school representatives modernize existing engineering controls and
coordinate with local authorities to allow them to become familiar with the school characteristics before an
event occurs7.
However, the outlined practices described by the Department of Homeland Security and MSA
Security Consultants are developed to serve more to the potential victims of the active shooter incident
and do not provide any recommendations about how responders shall enhance their methods for coping
with such a situation. In order to provide practical guidelines for responders proactively engaging in active
shootings at public schools, two necessary actions are required. First, the responders must have an
overview of these incidents and the involved subjects and be able to assess threats based on historical
and analytical data. The outcome of Dardsdale’s 2010 report8 greatly contributes towards the overall
active shooter threat assessment and can serve as a guideline in developing responders’ readiness.
Second, responders must be able to identify the effectiveness of a particular active shooter engagement
situation. Here, the analytical results from the modeling presented within this report can contribute to
identifying and improving responders’ methods and actions which are necessary for minimizing the
4
casualties of active shooting and maximizing school safety. Therefore, of each of the applied four
scenarios, it is the scenarios involving the employment of armed Resource Officers, faculty, or
combinations thereof, who are immediately available to react to an active shooter, that have been studied
least and makes these scenarios exceedingly viable. Further discussion and review of literature set forth
below examines these two particular categories in depth and provides validation for their use as scenario
conditions.
One of the given scenarios uses a limited number of concealed carry instructors (faculty or
employees) for a given location. The justification for employing this as a rational option is set forth below.
This option of introducing armed faculty is taken into consideration with both pro and anti-gun points of
view, including objections to this option from organizations such as the Brady Campaign to prevent gun
violence9, which disavow arming teachers and faculty. However, from an analysis of the literature and
practical point of view, the option of arming teachers and faculty remains credible with the researchers
and therefore exists as a realistic option in the methodology. Empirical data validating why the
introduction of firearms into the modeling scenarios is a viable option is set forth below.
In 2012 there were an estimated 1,214,462 violent crimes nationwide. This includes all violent
crime, including those in which firearms were used. This represents a decrease of over 12.9% from the
2008 level, a 15.4% decrease from the 2011 to 2007 level, and a 15.5% decrease from the 2011 to 2002
level10. At the same time firearms ownership increased sharply, by over 61%, or over 118 million
between 2004 and 201211. Additionally, during the timeframe of 1999-2000, a full 58% of firearms related
deaths were labeled as suicide, 38% as homicides and 3% ruled unintentional death by firearm12.
The Department of Justice commissioned a study in 1997 titled, Guns in America: National Survey on
Private Ownership and Use of Firearms. This study found the number of guns used in self-defense
annually at over 1.5 million13. This number exceeded the number of crimes in which a gun was used to
commit an act of violence. Additionally, since the tragic events at Sandy Hook School in 2012, a handful
of states have sought to restrict firearms, but 21 states have concretely expanded their firearms laws,
including many whose laws expanded opportunities for concealed carry holders to legally carry firearms
in previously restricted locations, including seven states now in which teachers or faculty in some schools
are armed14. Additionally, over 1300 pieces of legislation introduced nationwide since 2012 have
pertained to gun laws, with the majority of which seek to strengthen pro-gun laws and gun rights14. These
statistics indicate that a growing segment of educators, law enforcement personnel, and citizens are in
favor of either introduction armed security into schools or arming teachers themselves.
The evidence of growing firearm popularity and growing strength in both numbers, statistics relating to
crime and usage, and laws allowing their use create an undeniable data set that suggests that increased
firearms ownership and access does not contribute to increased crime, anecdotally, it statistically results
in a reduction10. As such, it remained as a valid option for analysis in the constructing of scenarios for
this study. Additionally, According to the National Center for Education Statistics, 57% of public schools
in the United States had no security staff present at any time during the week in 2009-2010, the most
recent year data were available. Even more nearly 70% had no police officer in the school every
week15.This data further compels the researchers to explore if incorporating this option as a variable in
the study could impact active shooter casualties.
Existing data on mass shooting events show overlying consistent themes such as location
chosen and time available2, 3, 4, 5, 6, 16, 17. Most of the mass shooting events have occurred in locations
such as schools, shopping malls, or other locations where people converge in masses2, 3, 4, 5, 6, 16, 17.
Although primary data for this research sought recent (past five year period) data, additional data
covering the most significant school shooting since 1966 was analyzed16,17. Analysis on these past
events with regard to casualties, location, and time of response are consistent with the interpretation
derived from the in-depth analysis of the five most recent; that being duration of event, location, and
5
ability for responders to act was critical in determining overall casualties. Almost all of these shootings
occurred in locations that are typically outside the scope of where licensed concealed carry holders are
permitted to carry weapons based on current laws18. A concealed carry law authorizes a citizen to
lawfully possess a firearm on or near their person in a concealed manner, or manner in which the weapon
is not readily visible from another. Examples are firearms kept in purses, in pockets, desk drawers or
vehicles.
Observing the mass shootings in schools, the environment can be comparted to a “closed
system” in which, despite the environment around it, the use, possession, or option of carrying a
concealed weapon is prohibited. This can be compared to, with justification for employing this
methodology for a scenario, larger environments, such as cities or even states. When looking specifically
at “crime spillover”, it becomes apparent how areas that allow for the carrying of concealed weapons
have decreased rates of crime compared to those which do not18. Additionally, the data supports the
conclusion that areas adjacent to those with concealed carry permits, and in turn do not authorize
concealed carry themselves, have higher rates of crime as criminals migrate to areas without concealed
carry in order to carry out criminal acts. This can be used in a microcosm view of schools or other likely
targeted locations for mass shootings. If schools are off limits to the carrying of concealed weapons, then
they therefore present themselves are a more lucrative target for mass shootings, just as cities who do
not possess concealed carry laws see larger amounts of crime if adjacent cities do permit the carrying of
concealed weapons.
Bronars and Lott’s study18 elaborates this phenomenon and employs the term, “geographic
spillover”. The authors study rates of crime over the timeframe of 1977 to 1992 across the demographic
spectrum of age, race, sex, income, welfare, and population density. The dependent variables used are
FBI uniform crime reports10 for the categories of violent crime, murder, rape, robbery, aggravated assault,
overall property crime, burglaries, auto thefts and larceny as reported per 100,000 population per county.
These factors were observed against the independent variable of concealed carry laws and arrest rates.
The stated objective was determining if shall-issue concealed weapons laws in one location alters crime
in neighboring adjacent areas. The authors posit that, taken as a whole, concealed carry laws (particularly
concealed handguns) do in fact deter criminals and that the greatest effect is seen when neighboring
counties adopt concealed carry policies. Their study concludes that locations on both the county and
state level are representative of the results noted. The authors further speculate that greatest overall
crime reduction can be achieved if concealed carry laws are permitted universally.
For the study, a neighboring county was defined as another geographic location with a center
within 50 miles of the studied county. To account for variations in arrests the study controls for violent or
property crime arrests depending on whether the crime rates studied are related to violence or property
crime. This mitigates the non-causal relationship between crime and arrest rates, as arrest rates are
functions of crime. The study states that the effects of “spillover” on a county without a concealed carry
when a neighboring county enacts a concealed carry law are substantial: an increase of 7.45% in rapes,
4.2% in robbery, and 4.5% in murder. These effects are insignificant if a neighboring county already has a
concealed carry law in place. When comparing crime rates of the county itself when implementing carry
laws, the rates of crime are reduced by an aggregate 34.16%. In all categories of crime except larceny,
the rates of crime are reduced over a seven year period by the adoption of concealed carry laws. In
studies where neighboring counties adopt concealed carry laws, and the host county already has
concealed carry laws, the only perceived effects are positive, or a decrease in all crime, except larceny18.
This therefore results in a significant increase in crime to areas without concealed carry laws when an
6
adjacent county implements such laws and no perceived increase in crime if the host country already
possesses such laws when neighboring counties, in turn, enact such legislation.
The article concludes through multiple examples of crime rate statistics that criminals tend to
migrate across areas with greater frequency when concealed carry laws are implemented. This migration
has a greater effect as related to concealed carry than just increased arrest rates, meaning increased law
enforcement techniques which lead to more arrests are still less effective at reducing crime than the
deterrent effect of having concealed carry laws. This spillover effect of crime is noted as immediate and
increased over time, with counties that implement such laws continually seeing a decrease in crime and
counties that don’t have concealed carry continually seeing a growth in crime. Taken as a whole, the
projection is that aggregate crime reduction can be better achieved through the adoption of concealed
carry laws in all states throughout the country18.
Again, the examples shown demonstrate not only what the effects of concealed carry are on
reducing crime in cities and states, but how adjacent cities and states who do not allow for concealed
carry see increased rates of crime. This translates, for this study, to schools or other locations
susceptible for mass shootings as these locations are comparative of “closed systems” in which crime is
more likely to migrate to as there is no immediate deterrent.
As outlined before, one of the effectiveness measures within the analysis is the response time of
the first responders. Regardless of the situation, the final determining factor in addressing mass shootings
is bringing in police and medical support in a timely manner. As illustrated by the example10, the “flash
to bang” factor, or ability for police to arrive in comparison to the start of a shooting event, directly relates
to the number of casualties inflicted. The study19 is based on data spanning a five year period and covers
24 school shootings in 18 states, and 41 workplace shootings in 12 states. The average time in shooting
events ranged from 3 to 4 minutes with an average victim being shot every 15 seconds. The fastest
police response time noted in these events was 5 to 6 minutes, with most taking much longer. Here, the
authors propose an armed responder, such as a resource officer or nearby law enforcement agent, as a
best option for reducing the severe outcomes of an active shooter incident.
In an example at Red Lake High School20, in Minneapolis, where a student killed five other
students, a security guard, and a teacher, the response of law enforcement was critical. Within two
minutes of the receiving the call, armed officers responded, headed toward the shooter and hit him twice
with gunfire. This caused the shooter to retreat from his position and commit suicide, preventing further
casualties. Overall the shooter’s attack lasted for over 10 minutes, but the quick response by law
enforcement ended the situation before further personnel were hurt.
Contrast this with situations such as the Virginia Tech School shooting in which the University’s
Police Department numbered over 35 officers, but the shooting events spanned a timeframe of over two
hours. When the shooter initially killed two personnel, improper procedures allowed for the campus to
remain unaware and the shooter was able to move undetected to another section of the campus and
begin shooting again. Even though police were present in mass numbers, they were fixated on the initial
shooting site and were unable to influence the second shooting site timely enough to prevent further
casualties21.
Multiple examples of active shooter incidents and the response time for law enforcement can
conclusively deduce that the longer an event transpires, the more casualties will be incurred.
Additionally, soft targets such as schools or other mass gatherings of people otherwise unable to defend
7
themselves make a more enticing target22. Additionally, the ability for first responders to arrive, organize,
and begin addressing the issue almost always results in reacting to the damage already done.
The increased likelihood of active shooter events has proven that even in areas with robust police
and military presence, the ability for active shooters to inflict mass damage quickly is not preventable with
external law enforcement or responders that must be called to the scene22. This implies that readily
available deterrents and responders, in the form of concealed carry personnel on scene have a greater
ability to end an active shooter situation sooner than waiting for law enforcement to arrive. Much of this
discussion focuses on select singular events. The situation becomes much more complicated when law
enforcement officers are forced to deal with multiple shooters or multiple locations. As Frazzano, 2010
stated, “Though smaller jurisdictions might have special tactics law enforcement squads, those squads
will not likely be able to deal with active shooter scenarios that include multiple shooters in multiple
locations with their own-source resources. How, then, are these jurisdictions to protect their citizens when
local capabilities and capacities are overwhelmed?”23 (p. 2)
In a recent study, the National School Security Task Force24 conducted an in-depth review of the
National Status of School Security. The study examined the history of school violence and offered
varying recommendations for decreasing violence in schools. The central point of the study referenced
the efficacy of having an armed first responder, such as a School Resource Officer (SRO) present. In the
study, the commission examined the effectiveness of a previous program sponsored in 1996 which
provided federal funding for school districts to conduct security evaluations and receive SRO
participation.
The program, sponsored by the U.S. Department of Justice was called COPS, Community
Oriented Policing Services and included a 60 million dollar, three-year grant to provide increased security
in the nation’s school systems. Although expired, the program provided valuable benefits and statistically
attributed to less crime during the timeframe in which it was implemented.
The study provided recommendations that included increasing the physical security of schools
and mental/behavioral health counseling to prevent and detect problem areas; increasing security through
either RSO or armed security of some form to include possible teacher/faculty arming. The overriding
consensus is that decreasing response time to threats and increasing ability for armed opposition to
engage an active shooter is the most important and effective method for reducing casualties24.
Process Flow Chart
The process flow chart for Scenario 1 is given in Figure 1. Since this a basic scenario, the model
will assume that no access control or any type of security is employed within the school. The active
shooter is assumed to be well armed and able to enter the school and randomly chooses the victims in
three potential areas: classrooms, common areas (cafeteria, library, gymnasium, etc…), or administration
offices. He can further randomly choose to change location and continue shooting in other areas until he
encounters a barrier (engaged by the law enforcement officers or commits suicide). Here, the response
time and the number of casualties and injured individuals will depend on the timeframe in which the
incident is reported and the response time of the law enforcement officers.
(Insert Figure 1)
The process flow chart for Scenario 2 is given in Figure 2. Here it is assumed that there is an
armed school resource officer present. The active shooter is assumed to be well armed and able to enter
8
the school and randomly chooses the victims in three potential areas: classrooms, common areas
(cafeteria, library, gymnasium, etc…), or administration offices. He can further randomly choose to
change location and continue shooting in other areas. This scenario assumes that once the shooter
begins his assault, the resource officer will act to mitigate the threat. Here, the response time and the
number of casualties and injured individuals will depend on the timeframe in which the incident is reported
and the response time of a barrier (the armed resource officer) can diffuse the situation, or confine it, until
law enforcement arrives.
(Insert Figure 2)
The process flow chart for Scenario 3 is given in Figure 3. Here it is assumed that there are 5%-
10% of employees (faculty and/or staff) exercising concealed carry. The active shooter is assumed to be
well armed and able to enter the school and randomly chooses the victims in three potential areas:
classrooms, common areas (cafeteria, library, gymnasium, etc…), or administration offices. He can
further randomly choose to change location and continue shooting in other areas. This scenario assumes
that staff and faculty with concealed carry will remain static in their respective locations and only respond
in a defensive posture to the threat, i.e. teachers with concealed carry would stay in their classrooms and
protect their students. Therefore their response is likely to be quantified through the data as less effective
than a resource officer who maneuvers to the threat. Here, the response time and the number of
casualties and injured individuals will depend on the timeframe in which the incident is reported and the
response time of a barrier (those individuals with concealed carry) can diffuse the situation, or confine it,
until law enforcement arrives.
(Insert Figure 3)
The process flow chart for Scenario 4 is given in Figure 4. Here it is assumed that there is an
armed school resource officer present in addition to 5%-10% of employees (faculty and/or staff)
exercising concealed carry. The active shooter is assumed to be well armed and able to enter the school
and randomly chooses the victims in three potential areas: classrooms, common areas (cafeteria, library,
gymnasium, etc…), or administration offices. He can further randomly choose to change location and
continue shooting in other areas. This scenario assumes that once the shooter begins his assault, the
resource officer will act to mitigate the threat by maneuvering to it, and those with concealed carry will
safeguard and defend from their current locations, thereby resulting in quicker incident culmination and
reduced casualties. Here the response time and the number of casualties and injured individuals will
depend on the timeframe in which the incident is reported and the response time of a barrier (the armed
resource officer/concealed carry personnel) can diffuse the situation, or confine it, until law enforcement
arrives.
(Insert Figure 4)
AnyLogic Model
Agent-based modeling is defined as “a system is modeled as a collection of autonomous
decision-making entities called agents. Agents may execute various behaviors appropriate for the system
they represent25. It is a form of computer simulation modeling that is becoming increasingly popular.
Borshchev, Karpov, and Kharitonov are experts in modeling software called AnyLogic26 and claim that
AnyLogic is one of the best pieces of agent-based modeling software in the world. It is widely used in
9
industry and academia. AnyLogic not only provides agent-based modeling capabilities, but it also allows
users to create discrete event and system dynamics models or even combinations of all three types.
Agent-based modeling was used to create the active shooter model in this research, and has
many benefits. It “captures emergent phenomena”, “provides a natural description of a system”, and “is
flexible”25. The agent-based modeling approach was chosen because it is the best technique for modeling
human systems. It allows the user to create complex interactions between humans, deal with people in a
limited amount of space, allows the population to be heterogeneous, allows the interactions to be
complex, and allows agents to execute complex behavior25. All five of these attributes are required in the
active shooter model.
Accurately creating a human agent-based model requires collecting the correct real-world data.
However, a limitation to this stems from the model only allowing a person to perform the predefined
actions that the user creates, and understanding that in reality humans possess free will27. This ultimately
results in model scenarios that replicate reality when provided with correct real-world data to great
efficacy, but never with total accuracy as the variable of free will remains undefined.
When the model is launched, the user is prompted with the model setup screen, shown in Figure
5.
(Insert Figure 5)
This screen allows the user to run the model with predefined inputs. The parameters to be
determined are the probability that teachers may have concealed carry weapons in their respective
classrooms and whether or not the school has an on-duty resource officer at the time of the incident. The
time for law enforcement to arrive and casualty rate are based upon the literature events previously
mentioned in the project description portion of this study. Once the parameters are set according to the
user’s preference, the user can click the button labeled “Run the model and switch to Main view.” This will
take the user to the Main view of the model and start the simulation.
Once the button is pressed, the Main view shows the floor plan of the school. The Main view is
shown in Figure 6. The walls have been traced with polylines using AnyLogic’s presentation pallet. This
serves as the environment for the agents to exist within.
(Insert Figure 6)
The active shooter appears at the front entrance of the school. If a resource officer is present, he
appears outside the doors of the gymnasium. The location of the active shooter and resource officer start
points can be changed using AnyLogic. The model runs in real time. Once it is completed, the results are
shown at the top. The results include how long responders took to engage and stop the shooter, how
many people were shot, and who the shooter was engaged by. An example of a result using the default
model settings is shown in Figure 7.
(Insert Figure 7)
The model works in three parts of logic. The first part is the active shooter and concealed
weapons carry logic, which is shown in Figure 8. The shooter enters through the front door of the school.
He then decides, at random, between one of five locations to start shooting. The five choices are Class1,
10
Class2, Class3, Office, and Cafeteria. The shooter, based upon reviewed literature, stays in the location
and shoots victims in 20-second intervals for two to five minutes before leaving and choosing another
destination. This will continue until the shooter is engaged and stopped. Only one stopping mechanic is
located within the active shooter logic. That is the chance that a teacher or a staff member has a
concealed weapon in the room which the shooter enters. If there is a person in the room with a concealed
weapon, the shooter is considered engaged, and the model is terminated.
(Insert Figure 8)
The second part of logic is the resource officer logic, which is shown in Figure 9. The resource
officer spawns at the predefined resource officer start point, which is currently the gymnasium door. He
then moves to a ready position in the hallway. Next, he is dispatched with the location of the shooter
inside the school. He moves to the location where the shooter was, unless the shooter has left the room.
If the shooter is still present, the resource officer engages the shooter and stops him. If the shooter has
already left, the resource officer stops and waits for the next location of the shooter. He then repeats the
process until he is able to engage the shooter.
(Insert Figure 9)
The third and final part of the model logic is the police logic, which is shown in Figure 10. It works
exactly like the resource officer logic with three exceptions. First, it passes multiple agents through the
logic (10 as of the time of this study). Second, the police enter through the front door of the school. Third,
police arrive several minutes after the shooting has already begun (5-20 minutes later as of the time of
this study). This is controlled using the discrete event framework shown in Figure 11. The police officers
start at the police station, or wherever they happen to be located at the time of the incident, and travel to
the school. Once at the school, they enter through the front doors and engage the shooter exactly as the
resource officer would.
(Insert Figures 10 & 11)
Results
Figures 12-17 show the results of all 50 runs for each of the proposed scenarios. Each graph
shows the number of casualties that occurred and the amount of time that passed between the shooter
entering the school and the time the shooter was stopped. A trendline is also present on each graph
showing a correlation between the number of casualties and the time to engage the shooter.
(Insert Figures 12-17)
Discussion
A compiled set of results is shown in Figure 18. These results include the average time to engage
and the average number of casualties calculated by the model in 50 runs of each scenario. As each
model run is random and independent, the scenario was run 50 times to ensure adequate sample size
would result in credible results. Scenarios 3 and 4 were split into two sub-categories, one with 5%
concealed carry and one with 10% concealed carry respectively.
(Insert Figure 18)
11
As seen in Figure 18, the number of casualties in all other scenarios is less than that of the basic
scenario. The comparison between having a resource officer and having teachers and staff with
concealed weapons shows that a resource officer is able to decrease casualties and response time more
effectively due to the resource officer being able to maneuver towards the threat while the teachers and
staff remain static. The effectiveness is most improved, however, when both a resource officer and
concealed carry personnel are present. Not surprisingly, increasing the percentage of concealed carry
personnel improved the response time and decreased the number of casualties.
Since the basic scenario showed the highest number of casualties, the other scenarios should all
be considered successful in minimizing the negative effect of active shooter phenomena. Having a
resource officer on duty reduced casualties by 66.4% and response time by 59.5%. Having 5% of
personnel carry a concealed weapon reduced casualties by 6.8% and response time by 5.4%. Increasing
the percentage of personnel with concealed carry to 10% reduced casualties by a total of 23.2% and
response time by 16.8%. Combining 5% concealed carry personnel with a resource officer reduced
casualties by 69.9% and response time by 59.7%. The final and most successful scenario of 10%
concealed carry personnel with a resource officer reduced casualties by 70.2% and response time by
62.7%.
The relationship between time to engage and number of casualties for each scenario is shown in
Figures 12-17. The trend lines confirm that, for each scenario, a longer response time has a positive
correlation with number of casualties.
Conclusion
The results of the study show that to decrease the number of casualties, the response time must
be reduced. The model data shows that the most efficient way to reduce response time is to have armed
personnel present at the school who can engage the active shooter before the police arrive. The
effectiveness of this method can be optimized by having both armed resource officers and armed
teachers or staff members with concealed weapons with which they can engage the shooter if he enters
their room. The results of this data can therefore be interpreted as when teachers and faculty serve as a
static deterrent or by not maneuvering on a shooter but rather just responding defensively, then the
greater the number of teachers or faculty armed, therefore result in a greater number of reduced
casualties.
Teachers and staff who choose to carry concealed weapons would need to be fully trained and
would likely be required to pass examinations to ensure that they are well-suited to carry concealed
weapons on school property. These examinations would likely be required multiple times throughout their
career. Very strict rules on where the weapons would have to be located would be needed. School
administrators would need to be willing to accept the liability of having weapons present in their schools.
Controversy exists over whether non-law enforcement personnel should be able to react to an
active shooter situation. Additional training of both law enforcement and concealed carry personnel would
be required to determine at what point self-defense measures transition to law enforcement roles.
Through additional training concealed carry personnel could maneuver towards active threats instead of
just sheltering in place. This, in conjunction with resource officers, would likely result in even fewer
casualties. However, considerations of friendly fire and liability issues preclude modeling this scenario at
this time as it assumes policy decisions. The results of the study show an improvement to both response
time and decreased number of casualties when responders are able to maneuver towards the threat.
Further research on the cost/benefit ratio of this topic should be done to determine whether the reduction
of casualties can be, or is, of value based on the training, casualties to students, and concealed carry.
Another area of future research would be to expand the model to recreate and analyze a historical event
12
to determine how concealed carry personnel and resource officers or law enforcement could have
mitigated the threat.
Lastly, it is the intent of the authors that rational discourse on the aforementioned topic will be
sought and reasonable alternatives to safeguard innocents from violence will be considered in the making
of policy decisions. A product of the research of active shootings in schools, and violence in general, is
the discussion of violence amongst youths. Fowler, 2009 conducted a study revealing that 50 to 96
percent of youth in urban environments are exposed to episodes of violence ranging from being a victim,
to witnessing or knowing first hand someone has been exposed to violent episodes. Over time, youth
exposed to violence increases the likelihood they will become victims of psychological disorders, such as
PTSD or insecurity. Fowler states these combined factors contribute to rising violence, especially among
young persons who are desensitized to violence and are therefore more prone to reacting with violence
themselves28.
Gil Kerlikowske, a staunch anti-gun advocate, concedes that addressing the issue of violence in
society by singling out guns alone will have little value. He reveals that starling levels of violence are
being identified in children, particularly those from fractured families or large urban settings29. As these
studies illustrate, there seems to be the distinct possibility of drawing a correlation to the rising violence
rate among youth, urbanization, and moral decay. This might also be substantiated as we look at the
historical context of the situation; firearms have been an intimate and substantial element of American
lifestyle since prior to the inception of the constitution, but it is only within the relative recent past that we
associate increased violence with access to guns. Therefore, this might suggest respective of firearms
being present, that changing culture, specifically that associated with urban development and changing
demographics, are more likely causal factors and indicators of violence, than firearms themselves. Even
in studies that control for social and economic factors, the results indicate that gun control does not
reduce violence or crime30. This suggests that despite best intentions and alternative efforts, the need to
arm school teachers or faculty for the defense of their students should not be dismissed on face value
simply because of the initial contemporary cultural aversion to firearms.
This data should compel us to look closely at the changing societal norms that seemingly produce
more young people with contempt for authority and less regard for life as a causal factor for many of the
incidents discussed in this report.
13
References
[1] Department of Homeland Security: Active Shooter: How to respond. Available at:
http://www.dhs.gov/xlibrary/assets/active_shooter_booklet.pdf. Accessed February 27, 2013.
[2] Office of the Virginia Governor: Report of the Virginia Tech Review Panel. Available at:
http://www.governor.virginia.gov/TempContent/techPanelReport.cfm. Accessed February 28,
2013.
[3] Northern Illinois University Board of Trustees: Report of the February 14, 2008 Shootings at
Northern Illinois University. Available at: http://www.niu.edu/feb14report/Feb14report.pdf.
Accessed February 28, 2013.
[4] NBC News Report: Official: 1 dead, 4 hurt in Ohio school shooting. Available at:
http://usnews.nbcnews.com/_news/2012/02/27/10517706-official-1-dead-4-hurt-in-ohio-school-
shooting. Accessed February 27, 2013.
[5] Wollan, M. Onishi, N.: Gunman Kills 7 in a Rampage at a Northern California University. Available at:
http://www.nytimes.com/2012/04/03/us/fatal-shootings-at-oikos-university-in-oakland-
calif.html?_r=1&. Accessed February 27, 2013
[6] Barron, J: Children Were All Shot Multiple Times With a Semiautomatic, Officials Say. Available at:
http://www.nytimes.com/2012/12/16/nyregion/gunman-kills-20-children-at-school-in-connecticut-
28-dead-in-all.html. Accessed February 27, 2013
[7] MSA Security: Special Analysis: The Active Shooter Thread. Available at:
http://www.msasecurity.net/Portals/91068/docs/MSA%20Special%20Analysis%20-
%20Active%20Shooter%20Threat%208.20.12.pdf. Accessed February 27, 2013
[8] Dardsdale, DA: Campus Attacks: Targeted Violence Affecting Institutions of Higher Education.
Available at: http://www2.ed.gov/admins/lead/safety/campus-attacks.pdf. Accessed February 27,
2013
[9] Brady Campaign to Prevent Gun Violence. Available at: http://www.bradycampaign.org/. Accessed
November, 20, 2013.
[10] Federal Bureau of Investigation Crime Reports. Available at http://www.fbi.gov/about-
us/cjis/ucr/ucr Accessed March 14, 2014.
[11] Krouse, William J: “Gun Control Legislation”, Congressional Research Report prepared for congress
November 14, 2012.
[12] Hemenway, David: “The Epidemiology of U.S. Firearm Injuries,” Journal of Public Health Policy;
2003; 24, 3/4.
[13] Cook, P.J., Ludwig, J: “Guns in America: National Survey on Private Ownership and Use of Firearms.”
National Institute of Justice, Research Brief, May, 1997.
[14] Law Center to Prevent Gun Violence. Retrieved from: http://smartgunlaws.org. Accessed 14
February, 2014.
[15] USA Today Report: NRA Plan Would Change Security in Most Schools. Available at:
http://www.usatoday.com/story/news/nation/2012/12/21/nra-newtown-police-in-
schools/1784953/. Accessed 14 February, 2014.
[16] Rostron, A., Siebel, B. (2007): No Gun Left Behind-The Gun Lobby's Campaign to Push Guns into
Colleges and Schools. Brady Center to Prevent Gun Violence.
14
[17] Smith, L., Supiano, B. (2007): Major shootings on American college campuses. Chronicle of Higher
Education, 53(34), A19-A19.
[18] Bronars SG, and Lott, J Jr.: Criminal deterrence, geographic spillovers, and the right to carry
concealed handguns. The American Economic Review, 88(2): 475-479.
[19] Buster C: Patrol response challenge. NCJRS Law and Order, 56(6): pp. 62-68.
[20] USA Today Report: Police response to Minn. shootings hailed. Available at:
http://usatoday30.usatoday.com/news/nation/2005-04-05-red-lake-response_x.htm#.Accessed
February 28, 2013.
[21] Droppings T: The Columbine resource office and the Virginia Tech police force. Available at:
http://www.tigerdroppings.com/rant/p/38637392/The-Columbine-resource-office-and-the-
Virginia-Tech-police-force.aspx.Accessed February 28, 2013.
[22] Rivera L: Active Shooter Incidents. Available at:
http://www.endesastres.org/files/Active_Shooter_Incident_FINAL.pdf.Accessed February 28, 2013.
[23] Frazzano T: Local jurisdictions and active shooters building networks, building capacities. Available
at: https://www.hsdl.org/?view&did=11541. Accessed February 28, 2013.
[24] National School Shield Task Force: Report of the National School Shield Task Force. Available at:
http://www.nraschoolshield.com/NSS_Final_FULL.pdf. Accessed February 28, 2013.
[25] Bonabeau E: Agent-based modeling: Methods and techniques for simulating human systems.
Avaliable at: http://www.pnas.org/content/99/suppl.3/7280.short#sec-1 .Accessed March 3, 2013.
[26] Borshchev, A, Karpov, Y, Kharitonov, V: From system dynamics and discrete event to practical agent
based modeling: Reasons, techniques, tools. Available at:
http://www2.econ.iastate.edu/tesfatsi/systemdyndiscreteeventabmcompared.borshchevfilippov04
.pdf. Accessed March 3, 2013.
[27] Muhdi, RA: Evacuation modeling: Development, characteristics, and limitations. Available at
http://www.mech.utah.edu/ergo/pages/NORA/2006/87-92%20Muhdi,%20Rani.pdf. Accessed
March 3, 2013.
[28] Fowler, P. J., et al.: “Community violence: A Meta-analysis on the Effect of Exposure and Mental
Health Outcomes of Children and Adolescents. Development and Psychopathology 21 (2009),
227259, 2009 Cambridge University Press, USA.
[29] Kerlikowske, G. R: “Safe at home? Policing the U.S. hometown in a post 9/11 Environment,” Journal
of Criminal Law and Criminology; July 2008; Vol. 50, 47-58.
[30] Moorhouse, J. C.,Wanner, B: “Does Gun Control Reduce Crime or does Crime Increase Gun
Control?” Cato Journal; Winter 2006; 26, 1.
15
Table 1. Active shooting incidents at educational institutions that took place in the last 5 years
Active Shooter Incidents
Location
Virginia Tech2
Northern Illinois
University3
Chardon High
School4
Oikos University 5
Sandy Hook
Elementary
School6
Date
7:15 am
9:51 am, April,
16, 2007
3:05 pm -3:11 pm
February 14, 2008
Approximately
7:30 am ,
February 27,
2012
Approximately
10:30 am, April 2,
2012
9:35 am 9:49
am, December 14,
2012
Target
Students and
faculty
Students and
faculty
School students
Stuff and random
students
Students and
staff
Shooter
Profile
23-year-old
Seung-Hui Cho,
a South Korean
citizen -
diagnosed with a
severe anxiety
disorder
Steven Phillip
Kazmierczak -
mental illness
Thomas M. Lane,
III - arrested
short time later in
a location outside
the school
One L. Goh -angry
at the
administration after
being expelled from
the university;
Surrendered after
siege
Adam Lanza -
diagnosed with
Asperger
syndrome
Number of
causalities
33 (including the
perpetrator)
6 (including the
perpetrator)
3
7
27 (including
perpetrator)
Number of
injured
23 (17 by
gunfire)
21 (17 from
gunfire)
3
3
2
Type of
weapons
Glock
19, Walther P22
12 gauge
Remington
Sportsman 48
shotgun;
9 mm; Glock 19
semiautomatic
pistol;
9mm Kurz Sig
Sauer P232
semiautomatic
pistol;
.380 Hi-Point
CF380
semiautomatic
pistol;
Ruger MK III .22
caliber semi-
automatic
handgun
.45-caliber
handgun with 10-
round magazines
223-caliber
Bushmaster
XM15-E2S rifle, a
10mm Glock
handgun and a
9mm SIG Sauer
P226 handgun
First
responder
actions
Police arrived
within three
minutes of
receiving an
emergency call
but took about
five minutes to
enter the
barricaded
building
Campus police on
scene within two
minutes of
shooting,
neutralized threat
within five min
The police
arrived quickly
and arrested the
shooter outside
of the school
(teacher was
chasing the
perpetrator)
n/a
Police arrive six
minutes after
shooting began
Disclaimer: Described work and the respective results given in this project report do not refer to any
particular incident or specific school location
16
Figure 1. Basic scenario of active shooting incident in a school
Figure 2. Active shooting incident in a school with resource officer.
17
Figure 3. Active shooting incident in a school with 5% - 10% concealed carry individuals
Figure 4. Active shooting incident in a school with 5% - 10% concealed carry and armed resource officer.
18
Figure 5. Model setup screen
19
Figure 6. Main view
Figure 7. Results displayed
20
Figure 8. Shooter and Concealed Weapons Carry logic
Figure 9. Resource officer logic
Figure 10. Police logic
21
Figure 11. Police travel logic
Figure 12. Basic scenario
Figure 13. Resource officer
0
5
10
15
20
25
30
35
0246810 12 14 16 18 20
Number of Casualties
Time to Engage (minutes)
0
5
10
15
20
25
30
35
0246810 12 14 16 18 20
Number of Casualties
Time to Engage (minutes)
22
Figure 14. 5% CCW
Figure 15. 10% CCW
0
5
10
15
20
25
30
35
0246810 12 14 16 18 20
Number of Casualties
Time to Engage (minutes)
0
5
10
15
20
25
30
35
0246810 12 14 16 18 20
Number of Casualties
Time to Engage (minutes)
23
Figure 16. 5% CCW + resource officer
Figure 17. 10% CCW + resource officer
0
5
10
15
20
25
30
35
0246810 12 14 16 18 20
Number of Casualties
Time to Engage (minutes)
0
5
10
15
20
25
30
35
0246810 12 14 16 18 20
Number of Casualties
Time to Engage (minutes)
24
Figure 18. Compiled results
0
2
4
6
8
10
12
14
16
18
20
Basic Resource
Officer 5% CCW 10 %
CCW 5% CCW
+
Resoruce
Officer
10 %
CCW +
Resoruce
Officer
Time (minutes)
and Number of Casualties (persons)
Scenario
Time to Engage (minutes)
Number of Casualties
... Xi and Chan, as well as Briggs and Kennedy have explored the risks and benefits of fighting back during an attack [7,8]. Others including Anklam et al. have considered the possibility of encouraging "concealed carry" of firearms on campuses [9]. There also exists multiple agent-based evacuation models; some of these incorporate psychological models of agent behavior (e.g. ...
... We set the report submission time as uniformly random between 1 and 3 minutes, and set the report processing time as uniformly random between 1 and 2 minutes, giving an overall time to initiate a security response as uniformly random between 2 and 5 minutes. This reflects the findings of Linger and Anklam et al. [4,9]. Due to a lack of empirical data, we set the probability that a student becomes aware of (or updates to the correct belief about) an attack happening at their location as 50%; through testing, we verified that this parameter value does not exert a significant primary effect on our variables of interest. ...
Full-text available
Article
Information sharing during emergency events often takes place through popular social messaging apps. An emergency response area of increasing concern in recent times is that of mass attacks on school campuses yet work studying information sharing systems in these situations is scarce. In this exploratory work, we used data from various sources to construct an agent-based simulation of a hypothetical knife attack on a university campus in southern China. We model the information sharing system after WeChat and evaluate the impact of group and individual messages on attack response and outcome. Evaluation metrics such as how students become alerted to the attack, the maintenance of accurate information, and casualty levels, are tracked. Results suggest that group messages may be most helpful despite being perceived as less trustworthy, and that the presence and structure of information sharing systems has a significant effect on response to campus attack events. Future work in the area is suggested.
... Other studies have focused on-campus attacks specifically. Anklam et al. (2015) have considered the possibility of encouraging ''concealed carry'' of firearms on campuses. Xi and Chan (2019) modeled the different dynamics and outcomes between a gun attack and a knife attack on the same campus. ...
Full-text available
Article
Information sharing systems are a critical component of emergency response—especially in campus attack situations that unfold very rapidly. The design of effective information sharing systems is often difficult, however, due to a lack of data on these assault events. This work takes an agent-based approach to simulate three campus emergency information sharing system design alternatives in the context of a college campus knife attack, and incorporates data from on-campus student surveys and parameter tuning experiments. Alternatives are evaluated according to: (1) improved student attack response outcomes; and (2) effective institutional response to the attack. The results confirm that increased awareness supports rapid emergency reporting, but an important gap exists between students’ awareness and their ability to respond effectively, which depends on a number of campus-specific factors. A strong positive impact is seen from safe and efficient information sharing with authorities. This impact depends largely on reporting system implementation qualities, as opposed to campus-specific factors. On a campus in China, WeChat was used as a basis for messaging models. The simulation results show a 9% drop in casualties and a 22% faster police response time from a text-based reporting system using “base” WeChat features instead of traditional phone reporting. Our results also project a 30% drop in casualties and 52% faster police response time using a system designed around a WeChat Mini Program or stand-alone campus emergency reporting app. These outcomes suggest a number of recommendations for improving outdated campus emergency information-sharing systems and response strategies.
... However, the effectiveness of armed individuals in stopping an active shooter has been questioned by researchers. While quick response time seems to be an important factor in stopping an active shooter (Anklam et al., 2014;Briggs & Kennedy, 2019), the effectiveness and accuracy of armed law enforcement in an active shooter situation has been called into question (Lewinski, Avery, Dysterheft, Dicks, & Bushey, 2015;Vickers & Lewinski, 2012). ...
Article
Proposals to arm teachers with guns have emerged in reaction to high-profile school shootings in the United States. Though controversial, these proposals have gained the support of various policymakers and special-interest groups. Public opinion, however, is largely divided on the issue and split on political lines. This study seeks to unpack the reasons why Americans may support arming teachers, drawing from multiple theoretical perspectives. Findings indicate support for the culture conflict perspective, with conservative respondents being more likely to support arming teachers and Black respondents and those who are more educated and have higher incomes less likely to support arming teachers. While instrumental concerns such as economic pessimism and anger at the country’s direction are associated with less support, those with children are more likely to support arming teachers. Findings on the role of feelings toward law enforcement on arming teachers are mixed. Implications for policy and research are discussed.
... This is accomplished in the model code, itself, as well as in published papers and model specification documentation. In the review of the extant computational models of mass shootings, for example, Briggs and Kennedy (2016) discovered one existing computational model that may have been overly optimistic in its assumptions: In this particular model, as soon as a simulated active shooter entered a room in which an armed responder was present, the shooter was immediately and completely neutralized merely as the result of the presence of an armed individual (Anklam, Kirby, Sharevski, & Dietz, 2015). Believing this assumption to be overly optimistic, Briggs and Kennedy (2016) consulted the peer-reviewed literature on police shooting performance and accuracy and developed and implemented an adjustable accuracy function in their model to account for the known variability in shooting performance, firearm effective ranges, and other circumstances that are known to affect a shooter's accuracy. ...
Full-text available
Article
Mass violence is empirically rare. Studying mass violence presents numerous meth-odological challenges. The complex nature of mass violence events, which may have germinated in years prior, make attempts to use conventional research methods problematic. Complexity science and the interdisciplinary field of computational social science offer new scientific paradigms and computational tools well suited to the study of complex and dynamic phenomena like mass violence. We review aspects of mass violence that can hamper research efforts, introduce complexity science, computational social science and computational modeling, and highlight three types of computational models that will likely be of particular interest and value to the threat assessment and management community. Public Significance Statement This theoretical review article discusses methodological challenges for mass violence research and proposes computational modeling and simulation as a valuable tool for use by threat assessment and management researchers and professionals. We discuss basic principles of complexity science, modeling, and simulation, and suggest three types of computational models-spatial/tactical, population, and organizational-of particular appeal for threat assessment and management. We conclude by presenting an example spatial/tactical agent-based model used to conduct computational research on the possibility of unarmed resistance in an active shooter scenario.
Article
Trained evacuation leaders in emergency offer the potential for improved decision making and evacuation. Compared to victims, trained evacuation leaders can make educated assessments of the situation based on their training, knowledge of the facilities, and additional details about the incident, which enables them to guide victims in choosing a safe departure time and evacuation route. Despite a general understanding about the benefits of such leaders in evacuation, mass shooting cases require a separate attention because these cases are more complex with different behavioral decisions, not just running away, with a continuously changing source of the hazard source, the shooter. This study develops a simulation model package and evaluates the effect of trained evaluation leaders on the victim safety during an active shooter incident. The study leverages sophisticated human motion dynamics models and human behaviors supported by past literature in an agent-based model. The study varies several parameters (e.g., occupancy, firing rates and gun range, and victims’ decision of running or hiding) in this simulation to draw generalized conclusions on the leaders’ impact on various scenarios. The results reveal general findings with several interesting points. Overall, increased leaders’ presence contributes to fewer fatalities. Even few trained leaders, compared with none, can considerably improve victim safety. Even if leaders are not uniformly positioned, they still provide substantial benefits for victim’s safety. The leaders’ benefits were consistently found in various parametric studies (e.g., number of leaders, occupancy, leaders’ strategic placement, gun range, and shooting rate) that support the mentioned findings.
Chapter
This chapter examines weapon carrying by teachers and students in K-12 schools. Most of the chapter focuses on juveniles, exploring how often juveniles bring weapons to school, the demographic characteristics these juveniles have in common, and why juveniles opt to bring weapons on school grounds. Empirical studies of these topics based on official data, victimization data, and self-report data are reviewed. Additionally, this chapter details the federal and state laws that regulate weapons on school property. Attention is also given to the recent debate over arming K-12 teachers and staff as a prevention measure. Like the discussion of juvenile gun carrying, this chapter highlights the law regulating this issue, arguments for and against such a policy, and the empirical research assessing effects.
Chapter
This chapter examines weapon carrying by teachers and students in K-12 schools. Most of the chapter focuses on juveniles, exploring how often juveniles bring weapons to school, the demographic characteristics these juveniles have in common, and why juveniles opt to bring weapons on school grounds. Empirical studies of these topics based on official data, victimization data, and self-report data are reviewed. Additionally, this chapter details the federal and state laws that regulate weapons on school property. Attention is also given to the recent debate over arming K-12 teachers and staff as a prevention measure. Like the discussion of juvenile gun carrying, this chapter highlights the law regulating this issue, arguments for and against such a policy, and the empirical research assessing effects.
Article
Active shooting (AS) violence poses a serious threat to public safety. Most importantly, this number is increasing at an alarming rate over the last few years. Unfortunately, the knowledge base on attack detection and civilian response strategy under an AS incident are very limited. Such a lack of understanding hinders the development of effective training modules, response, and mitigation strategies (for both civilians and law enforcement officers) for an event that is highly unpredictable and unfolds quickly (typically in less than 5 min). To fill this gap in the literature, this article proposes an innovative mathematical model which captures several important features (e.g., capacity of the facility and individual choices, heterogeneity of individual behavioral and choice sets, restriction on choice sets depending on the location of the shooter and facility orientation, and many others) which are essential for appropriately characterizing and analyzing the response strategy for civilians under an AS exposed environment. We demonstrate the applicability of the proposed model by implementing the effectiveness of the RUN.HIDE.FIGHT. program in an academic environment. Results indicate that cognitive delay, the initial distribution of the civilians, and building configurations (e.g., location and number of entrances/exits) play a significant role in the safety of civilians under an AS incident.
Article
This study investigated the role of workplace preparedness actions in employee perceptions of workplace risk, workplace preparedness, and personal self-efficacy in an active shooter event. Data were drawn from an online, state representative survey of 668 Pennsylvania residents in 2019. Nearly 40% of employees reported their workplaces had not taken any preparedness actions. Having a workplace take a greater number of preparedness actions was associated with increased self-efficacy and increased perceptions of workplace preparedness, but also an increase in perceived risk. Males and gun owners perceived lower levels of workplace risk and reported substantially higher self-efficacy. However, associations between workplace efforts and self-efficacy differed from those for perceived workplace preparedness. Associations with firearm policy and the presence of security staff also differed for the two outcomes.
Full-text available
Article
On April 16, 2007, the day of the "Massacre at Virginia Tech", in which 32 innocent college students and faculty lost their lives to a crazed gunman armed with two semi-automatic pistols and a couple hundred rounds of ammunition, the first reaction of the gun lobby was that we need more guns on the college campuses of our Nation. The gun lobby also wants to repeal the Federal Gun-Free School Zones Act and arm public school teachers. This report exposes the gun lobby's step-by-step strategy to force guns into every aspect of daily life. Introducing guns into schools and universities is only the latest attempt to knock down barriers to firearms possession in places that previously were gun-free. Section One discusses the severe risks that would be created if gun possession and carrying became widespread on college campuses and gun-free school zones were undermined. Section Two explains how that campaign would destroy fundamental rights of academic freedom and wrest control of college campuses and schools from persons entrusted to secure those institutions. Section Three explains that these risks can best be managed by continuing gunfree policies. The Conclusion stresses that because of the gun lobby's campaign, it is no longer enough for academic institutions to adopt gun-free policies. Academic communities must also become active in potentially every state legislature if they want to keep the right to maintain a gun-free environment.
Full-text available
Article
Meta-analytic techniques were used to estimate the effects of exposure to community violence on mental health outcomes across 114 studies. Community violence had its strongest effects on posttraumatic stress disorder (PTSD) and externalizing problems and smallest impact on other internalizing symptoms. Victimization by community violence most predicted symptomatology compared to witnessing or hearing about community violence. Witnessing community violence had a greater effect than hearing about violence on externalizing problems, but both types of exposure had an equal impact on other internalizing problems. PTSD symptoms were equally predicted by victimization, witnessing, or hearing about community violence. Compared to children, adolescents reported a stronger relationship between externalizing behaviors and exposure, whereas children exhibited greater internalizing problems than did adolescents.
Article
Congress has debated the efficacy and constitutionality of federal regulation of firearms and ammunition, with strong advocates arguing for and against greater gun control. Since March 2011, much of the gun control debate in the 112th Congress has swirled around allegations that the Department of Justice (DOJ) and the Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF) mishandled a Phoenix, AZ-based gun trafficking investigation known as "Operation Fast and Furious." In the Consolidated and Further Continuing Appropriations Act, 2012 (P.L. 112-55), Congress included a provision that reflects a Senate-adopted amendment that forbids the expenditure of any funding provided under it to be used by a federal law enforcement officer to transfer an operable firearm to a person known or suspected to be connected with a drug cartel without that firearm being continuously monitored or controlled. The act, however, does not include language adopted during House full committee markup to prohibit ATF from collecting multiple rifle sales reports in Southwest border states. On November 16, 2011, the House passed a bill (H.R. 822) that would establish a greater degree of reciprocity between states that issue concealed carry permits for handguns to civilians than currently exists under state law. On October 11, 2011, the House passed a Veterans' Benefits Act (H.R. 2349) that would prohibit the Department of Veterans Affairs from determining a beneficiary to be mentally incompetent for the purposes of gun control, unless such a determination were made by a judge, magistrate, or other judicial authority based upon a finding that the beneficiary posed a danger to himself or others. In May 2011, firearms-related amendments to bills reauthorizing the USA PATRIOT Act were considered (H.R. 1800, S. 1038, and S. 990), but they were not passed. The tragic shootings in Tucson, AZ, on January 8, 2011, in which six people were killed and 13 wounded, including Representative Gabrielle Giffords, have also generated attention. Several members introduced proposals that arguably address issues related to the shooter's mental illness and drug use (see S. 436/H.R. 1781) and his use of large capacity ammunition feeding devices (LCAFDs) (see H.R. 308 and S. 32), as well as a proposal to ban firearms within the proximity of certain high-level federal officials (see H.R. 367 and H.R. 496). This report concludes with discussion of other salient and recurring gun control issues that have generated past congressional interest. Those issues include (1) screening firearms background check applicants against terrorist watch lists, (2) reforming the regulation of federally licensed gun dealers, (3) requiring background checks for private firearms transfers at gun shows, (4) more-strictly regulating certain firearms previously defined in statute as "semiautomatic assault weapons," and (5) banning or requiring the registration of certain long-range .50 caliber rifles, which are commonly referred to as "sniper" rifles. To set these and other emerging issues in context, this report provides basic firearms-related statistics, an overview of federal firearms law, and a summary of legislative action in the 111th and 112th Congresses.
Article
The emerging threat of terrorism, specifically small unit active shooter attacks, is a concern for American law enforcement agencies. Events like Mumbai (November 26, 2008) and Beslan (September 1, 2004) demonstrate the vulnerability of local law enforcement officers in defending against multiple attackers and multiple locations. Smaller jurisdictions (populations less than 100,000 people) are challenged with administrative and operational capacities considerably more than larger jurisdictions. Therefore, smaller agencies must find ways to enhance their capacities within tightening budgetary constraints. To investigate this problem and find workable solutions, qualitative research methods of case studies and interviews were employed. Specifically, Mumbai, Beslan and two high-profile United States incidents (Columbine High School shooting (April 20, 1999) and North Hollywood Bank shoot out (February 28, 1997) were studied. Individuals from the U.S. cases were interviewed to explore information not necessarily documented. Data from the case studies and interviews were collated and reviewed for common themes. These themes were analyzed to draw conclusions on how smaller jurisdictions should proceed in building capacities to deal with active shooter scenarios. Findings suggest that smaller jurisdictions can build capacities by creating a megacommunity within local law enforcement. This includes developing systems to share smart practices, training for small unit attacks, and creating multi-jurisdictional interoperability standards.
Conference Paper
This paper may be considered as a practical reference for those who wish to add (now sufficiently matured) Agent Based modeling to their analysis toolkit and may or may not have some System Dynamics or Discrete Event modeling background. We focus on systems that contain large numbers of active objects (people, business units, animals, vehicles, or even things like projects, stocks, products, etc. that have timing, event ordering or other kind of individual behavior associated with them). We compare the three major paradigms in simulation modeling: System Dynamics, Discrete Event and Agent Based Modeling with respect to how they approach such systems. We show in detail how an Agent Based model can be built from an existing System Dynamics or a Discrete Event model and then show how easily it can be further enhanced to capture much more complicated behavior, dependencies and interactions thus providing for deeper insight in the system being modeled. Commonly understood examples are used throughout the paper; all models are specified in the visual language supported by AnyLogicTM tool. We view and present Agent Based modeling not as a substitution to older modeling paradigms but as a useful add-on that can be efficiently combined with System Dynamics and Discrete Event modeling. Several multi-paradigm model architectures are suggested.
Article
As buildings and structures are designed more creatively, evacuation procedures to assure acceptable building fire safety standards have been a major concern of fire safety professionals. One way to assure occupants' safety lies in evacuation drills. Another potential alternative to confidently evaluate occupants' safety in buildings lies in computer-based evacuation models. The basis on which evacuation models have been developed are discussed. Human performance and behavior research during evacuation are also presented. The shortcomings of current evacuation models and the need for new features and capabilities are suggested for more accurate and representative evacuation models.
Article
Despite the prominence of terrorism concerns on the national agenda, three areas of public policy pose more significant challenges for local law enforcement in the United States: illegal gun proliferation and distribution; offender incarceration and re-entry paths; and investments in the lives of children. This paper argues that the current direction of public policy in these latter areas should be a primary concern, because these policies not only impact law enforcement in a negative way, but also threaten the strength and vitality of the communities law enforcement is trying to serve. Moreover, the dangers posed by these policies are far more calculable, more likely and more destructive over the long run than those posited for more extreme, though less-likely threats, to which the U.S., as a nation, is committing enormous sums of money, for seemingly incremental, public safety benefits. The role of law enforcement executives is critical to how these issues will be addressed.