Content uploaded by Mark Gius
Author content
All content in this area was uploaded by Mark Gius on Feb 26, 2018
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=rael20
Download by: [Quinnipiac University] Date: 19 April 2017, At: 08:00
Applied Economics Letters
ISSN: 1350-4851 (Print) 1466-4291 (Online) Journal homepage: http://www.tandfonline.com/loi/rael20
The effects of state and Federal gun control laws
on school shootings
Mark Gius
To cite this article: Mark Gius (2017): The effects of state and Federal gun control laws on school
shootings, Applied Economics Letters, DOI: 10.1080/13504851.2017.1319555
To link to this article: http://dx.doi.org/10.1080/13504851.2017.1319555
Published online: 19 Apr 2017.
Submit your article to this journal
View related articles
View Crossmark data
ARTICLE
The effects of state and Federal gun control laws on school shootings
Mark Gius
Quinnipiac University, Hamden, CT, USA
ABSTRACT
School shootings are the highest profile type of murder in the United States. They are also the
rarest type of murder. In 2014, there were only 17 firearm murders that were perpetrated in
schools and colleges. The purpose of the present study is to determine the relationship between
school shootings and state and Federal gun control laws. Using a Poisson, two-way fixed effects
model, it was found that assault weapons bans reduced the number of school shooting victims
by 54.4%. All other gun control laws (concealed carry laws, private sale background checks and
Federal dealer background checks) had no statistically significant effects on school shootings.
Although assault weapons bans may reduce the overall number of school shooting victims, the
average reduction in murder victims may be less than 10 per year. Hence, it is unclear if gun
control is the most appropriate policy to use to reduce the number school shooting victims.
KEYWORDS
Gun control; school
shootings; state laws;
Federal laws
JEL CLASSIFICATION
K40; K14
1. Introduction
Although relatively rare, school shootings are among
the highest profile type of murder in the United States.
In 2014, there were 14 249 murders, 17 of which were
firearm murders that were perpetrated in schools and
colleges. As can be ascertained from Figures 1 and 2,
the number of persons killed and injured by guns in
schools varies dramatically from year to year, although
it does appear as if both injuries and deaths resulting
from school shootings are trending upwards.
Shortly after most school shootings, various pub-
lic policy measures that may reduce the frequency of
these shootings are suggested. Most involve some
type of gun control measure. One problem with
examining the effects of gun control, or any type of
public policy, on school shootings is the relative
rarity of such events. In most years, fewer than 20
people nationwide are shot and killed in a school
setting. The infrequency of these events is one rea-
son why there is so little empirical research on the
effects of gun control on school shootings.
One of the few studies that has examined the
effects of gun control laws on school shootings is
Kalesan et al. (2016). In this study, the authors
examined the relationship between background
checks for firearm and ammunition purchases and
school shooting incidents. Using state-level data for
the period 2013–2015 and various control variables,
the authors found that states with background
checks had lower incidents of school shootings. In
addition, states with higher levels of mental health
spending and educational spending also had lower
rates of school shootings.
There are several issues with this study, however.
First, all school shooting incidents were included in
their analyses, even when no one was killed or
injured. The authors admit that when they excluded
those incidents with no killings, the statistical sig-
nificance of their results was reduced. Second, their
measure of gun ownership is not universally
accepted and is based on a survey conducted by
one of the authors of the study. A more appropriate
measure of the prevalence of gun ownership would
have been to use one of the more established surveys
that examine gun ownership, such as the General
Social Surveys, or to use a proxy measure, such as
the proportion of suicides that are committed using
a firearm. Third, the use of a K-12 educational
spending variable is circumspect, especially given
that they examined shootings in all educational set-
tings, including shootings that occurred at colleges
and universities.
Hence, even though the methodology used in
Kalesan et al. (2016) is questionable, it is one of the
CONTACT Mark Gius Mark.gius@quinnipiac.edu Quinnipiac University, Hamden, CT 06518, USA
APPLIED ECONOMICS LETTERS, 2017
http://dx.doi.org/10.1080/13504851.2017.1319555
© 2017 Informa UK Limited, trading as Taylor & Francis Group
few studies that have examined the relationship
between school shootings and gun control laws. In
order to address that deficiency in the area of gun
control research, the present study will attempt to
determine the relationship between school shootings
and state and Federal gun control laws. This study
differs from prior research in several ways. First, a
much larger data set will be used; state-level data for
the period 1990–2014 will be analysed. Second, only
school shootings that resulted in an injury or death
will be included in the data. Finally, both state and
Federal gun control laws will be examined.
2. Empirical technique
Using Gius (2015)as a guide, the following equation
was estimated in the present study:
Y¼α0þα1State Assault Weapons Ban
þα2State Background Checks
þα3State Concealed Weapons Laws
þα4Federal Background Checks
þα5Control Variables þα6State Fixed Effects
þα7Year Fixed Effects
(1)
where Yis the number of deaths and injuries due to
school shootings. Control variables include the follow-
ing: population density; percentage of population that
has a 4-year college degree; per capita median income;
annual unemployment rate; percentage of population
that is ages 5–18; per capita alcohol consumption; and
the proportion of suicides that are firearm suicides. The
firearm suicide variable is used as a proxy for gun own-
ership prevalence (Lang 2013).
0
5
10
15
20
25
30
35
40
45
50
Figure 1. Number of persons killed in school shooting: 1990–2014.
0
5
10
15
20
25
30
35
40
45
50
Wounded
Figure 2. Number of persons wounded in school shooting: 1990–2014.
2M. GIUS
The following Federal and state gun control laws
are examined in the present study: Federal assault
weapons ban (1994–2004); Federal background
checks for gun purchases from dealers (1994 to pre-
sent); state-level assault weapons ban; state-level
restrictive concealed carry laws; and state-level back-
ground checks for private party sales. The state
assault weapons ban variable is expressed as a
dummy variable that equals 1 if the state had an
assault weapons ban and 0 otherwise. Since the
Federal assault weapons ban was in effect during
the period 1994–2004, it was felt that the most
appropriate way to incorporate that law into the
analysis would be to set the assault weapons ban
dummy variable to 1 for all states during the
Federal assault weapons ban period. Combining
these similar laws into one variable should mitigate
any issues that may have arisen due to multicolli-
nearity or to misspecification of the model.
Concealed carry weapons (CCW) laws deal with
how permits are issued to individuals who want to
carry concealed weapons. The most restrictive types
of CCW laws are ‘may issue’and prohibited. In a
‘may issue’state, local and state authorities can deny
requests for concealed carry permits, even requests
from qualified applicants. May issue CCW laws are
considered restrictive. ‘Prohibited’states do not
allow private citizens to carry concealed weapons.
In the present study, the CCW dummy variable
equals 1 if the state was may issue or prohibited and
0 otherwise.
Finally, both the Federal background check law
(Brady Act) and state-level private sales background
check laws are used in the present study. The Brady
Act only imposed background check requirements on
firearm purchases made from Federally licenced firearm
dealers. Private party firearm sales are exempt from
Federal background checks. Given that the Federal back-
ground check law went into effect in 1994, the Federal
Background check dummy variable equals 1 for the
period 1994 to present and 0 otherwise. Regarding
state-level background checks, only state-level laws
requiring background checks for private sales of firearm
are included in the model. If a state requires a back-
ground check for any type of private sale, then the
private sale background check dummy variable equals 1.
All data used in the present study are state-level
and were obtained for the years 1990–2014. Socio-
economic data were obtained from the Statistical
Abstract of the United States and other relevant
Census Bureau documents. Data on alcohol con-
sumption were obtained from the National
Institute on Alcohol Abuse and Alcoholism.
Information on state-level gun control laws were
obtained from Ludwig and Cook (2003), the Law
Center to Prevent Gun Violence and the National
Rifle Association.
Data on school shootings were obtained from
Kalesan et al. (2016), Klein (2012) and Everytown
USA. According to this data, there were a total of
382 people injured and 354 people killed in school
shootings during the period 1990–2014. The average
number of school shooting fatalities per year was 14,
and the average number of school shooting injuries
per year was 15.
3. Results and concluding remarks
Equation 1 was estimated using a Poisson, two-way
fixed effects model, controlling for both state and
year fixed effects. All observations were weighted by
state population. A Poisson model was used because
the dependent variable is count data (the number of
school shooting victims). Results are presented in
Table 1.
According to these results, the only gun control
measure that had a statistically significant effect on
the number of school shooting victims was the
assault weapons ban. When the assault weapons
ban (state or Federal) was in effect, the number of
school shooting victims was 54.4% less than when
the assault weapons ban was not in effect, holding all
other factors constant. This result is corroborated by
the data presented in Figure 1. As can be seen from
the chart, the number of persons killed in school
Table 1. Poisson fixed effects regression results.
Variable Coefficient Test statistic
Assault Weapons Ban −0.785 −4.81***
Federal Background Checks 0.346 1.45
State Background Checks 0.223 0.94
Concealed Carry Laws −0.185 −1.08
Population density 0.028 4.84***
Per capita median income 0.00007 4.29***
Proportion of population with college degree 0.596 0.33
Unemployment rate 10.054 3.94***
Proportion of population aged 5–18 22.25 3.45***
Per capita alcohol consumption 2.041 4.70***
Ratio of firearm suicides to total suicides 13.703 8.58***
Note: The incidence rate ratio can be obtained by exponentiating the
coefficient.
***p-value <1%.
APPLIED ECONOMICS LETTERS 3
shootings was relatively low and stable until 2004,
when the Federal assault weapons ban expired. After
2004, the number of persons killed in school shoot-
ings increased dramatically and became much more
volatile. All other gun control laws (CCW laws,
private sale background checks and Federal dealer
background checks) had no statistically-significant
effects on the number of school shooting victims.
Regarding the significance of the control vari-
ables, the unemployment rate, per capita alcohol
consumption, median household income, the per-
centage of the population between the ages of 5
and 18, population density and the ratio of firearm
suicides to total suicides (a proxy for the prevalence
of gun ownership) were all significantly and posi-
tively related to the number of school shooting vic-
tims. Hence, those states that had above-average
unemployment and gun ownership rates along with
greater than average alcohol consumption and larger
student-age populations had greater numbers of
school shooting victims. Some of these control vari-
ables had much greater effects on the number of
shooting victims than did the assault weapons ban.
School shootings are among the most horrific of
crimes. Although they are a very small share of
overall murders, they typically capture the atten-
tion of the entire nation. Places such as Sandy
Hook and Columbine recall terrible moments in
American history. One impediment to developing
public policies that may reduce the number of
school shootings, however, is the relative infre-
quency of these shootings.
Gun control is one of the more commonly pro-
posed policies in response to school shootings.
Unfortunately, as this study has shown, most gun
control policies have no significant effects on school
shootings. Although assault weapons bans may
reduce the overall number of school shooting vic-
tims, the average reduction in fatalities may be less
than 10 per year. Given these results, it is unclear if
gun control is the most appropriate policy to use to
reduce the incidence and severity of school
shootings.
Disclosure statement
No potential conflict of interest was reported by the author.
References
Gius, M. 2015.“The Impact of State and Federal Assault
Weapons Bans on Public Mass Shootings.”Applied
Economics Letters 22 (4): 281–284. doi:10.1080/
13504851.2014.939367.
Kalesan,B.,K.Lagast,M.Villarreal,E.Pino,J.Fagan,andS.
Galea. 2016.“School Shootings during 2013-2015 in the USA.”
Injury Prevention.doi:10.1136/injuryprev-2016-042162.
Klein, J. 2012.The Bully Society: School Shootings and the
Crisis of Bullying in America’s Schools. New York, NY:
NYU Press.
Lang, M. 2013.“Firearm Background Checks and Suicide.”
The Economic Journal 123: 1085–1099. doi:10.1111/
ecoj.2013.123.issue-573.
Ludwig, J., and P. Cook, eds. 2003.Evaluating Gun Policy:
Effects on Crime and Violence. Washington, D.C.: The
Brookings Institution.
4M. GIUS
