Content uploaded by Pontus Leander
Author content
All content in this area was uploaded by Pontus Leander on Mar 22, 2018
Content may be subject to copyright.
https://doi.org/10.1177/0146167217703952
Personality and Social
Psychology Bulletin
1 –15
© 2017 by the Society for Personality
and Social Psychology, Inc
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/0146167217703952
pspb.sagepub.com
Article
Wickenburg, Arizona, was recently ranked among the safest
small towns in Arizona (ValuePenguin, 2015). Yet here is an
excerpt of an interview with a local gun owner about his rea-
sons for carrying daily:
It’s a situation just like getting up in the morning and putting
your shoes on or your boots on . . . I’m carrying it because I want
people to know that if there is an incident somewhere, that I am
there to defend myself or my family. (Hwang & Murphy, 2014)
There are an estimated 270 to 310 million guns privately
owned in the United States, which makes Americans the
world’s best armed people (GunPolicy.org, 2016). A recent
Gallup Poll suggests that the main reason for owning a gun—
mentioned by 60% of gun owners—is protection/self-defense
(Gallup, 2014). From a psychological perspective, it seems
likely that defensive gun ownership is motivated by subjec-
tive factors such as perceived risk of victimization rather
than a person’s objective risk of attack. Yet research and
theorizing on the need for guns—the reasons for owning
them—have been almost completely neglected by psycholo-
gists. The need for theory in this area is emphasized by crim-
inologists, who suggest that, “Understanding the factors that
lead people to obtain guns for self-protection is important for
both theoretical and policy reasons” (Kleck, Kovandzik,
Saber, & Hauser, 2011, p. 312); they also note, “Studies
assessing the effect of fear/risk and criminal victimization on
gun ownership have obtained wildly varying results”
(p. 313). Yet some fears and perceived risks are so diffuse or
generalized in one’s mind that they do not necessarily pertain
to any one specific crime in particular and may thus be more
subjective than objective per se. Although psychologists
have long considered the role of perceived threat in moti-
vated behavior and reasoning, they have yet to apply it to the
psychology of gun ownership.
In this article, we develop and test the first psychological
theory of the motivational bases of gun ownership. There are
three reasons why a psychological theory is needed. First, as
we will argue, there appears to be little objective justification
for this prevalent need of a gun for self-defense. Second, as
703952PSPXXX10.1177/0146167217703952Personality and Social Psychology BulletinStroebe et al.
research-article2017
1University of Groningen, The Netherlands
2University of Maryland, College Park, USA
Corresponding Author:
Wolfgang Stroebe, Department of Social and Organizational Psychology,
University of Groningen, Grote Kruisstraat 2/1, 9712TS Groningen, The
Netherlands.
Email: wolfgang.stroebe@gmail.com
Is It a Dangerous World Out There?:
The Motivational Bases of American
Gun Ownership
Wolfgang Stroebe1, N. Pontus Leander1,
and Arie W. Kruglanski2
Abstract
Americans are the world’s best armed citizens and public polling suggests protection/self-defense is their main reason for
gun ownership. However, there is virtually no psychological research on gun ownership. The present article develops the
first psychological process model of defensive gun ownership—specifically, a two-component model that considers both the
antecedents and consequences of owning a gun for protection/self-defense. We demonstrate that different levels of threat
construal—the specific perceived threat of assault and a diffuse threat of a dangerous world—independently predict handgun
ownership; we also show how utility judgments can explain the motivated reasoning that drives beliefs about gun rights. We
tested our model in two independent samples of gun owners (total N = 899), from just before and after the Orlando mass
shooting. This study illustrates how social-cognitive theories can help explain what motivates Americans to own handguns
and advocate for broad rights to carry and use them.
Keywords
motivated cognition, gun-related beliefs, threat perceptions, self-defense motives
Received August 30, 2016; revision accepted March 14, 2017
2 Personality and Social Psychology Bulletin
public health researchers have demonstrated repeatedly,
there is little evidence that gun owners are able to use their
guns in situations that might demand and justify defensive
gun use. A third reason is that there is a fierce debate over the
proper role of guns in American society, yet there exists vir-
tually no explanation for what motivates some Americans to
advocate for broad gun rights under the Second Amendment
of the Constitution.
Our goal is to provide a starting point for readers to con-
sider American gun ownership to be a social-psychological
phenomenon, wherein our theories can be applied in a mean-
ingful, societally relevant way. Thus, after we provide a brief
primer on the current literature on American gun ownership,
we develop a motivational model of defensive gun owner-
ship. We take a social-cognitive perspective to explain how
two components—antecedent perceptions of threat and con-
sequent utility judgments—can predict the motivation that
leads to the purchase of a handgun (a defensive firearm) and
also endorsement of broad rights to carry and use one’s gun.
The antecedent component pertains to the various types of
threat that may trigger a goal for protection/self-defense; the
consequent component pertains to the utility judgments that
motivate ownership of defensive firearms and a desire to
maximize one’s affordances to obtain, carry, and use them.
Our model is rooted in the idea that the need for safety
and security is fundamental and that the mere perception of
threat can frustrate this need. This in turn activates a search
for a means of protection/self-defense. In American culture,
the gun has a special place in this regard—It is enshrined in
the Constitution and made salient in popular culture. Thus,
higher levels of perceived threat may intensify beliefs that
gun ownership and availability are important. We test our
two-component model via surveys of male gun owners and
nonowners.
Are Guns Used—or Useful—for
Self-Defense in the United States?
Why do millions of Americans feel the need to own a gun for
protection/self-defense? The most common theoretical
explanation from criminologists is a specific “fear of crime”
or “perceived risk of victimization” (Cao, Cullen, & Link,
1997; DeJong, 1997; Kleck et al., 2011; Williams & McGrath,
1976). Yet there appears to be little relationship between sub-
jective and objective risk, which makes it difficult to under-
stand the widespread need for guns for protection/
self-defense. Homicide rates in the United States have
decreased over the last few decades, to a rate of 4.5 per
100,000 people killed (Federal Bureau of Investigation
[FBI], 2014). Furthermore, if gun ownership were purely
driven by the objective risk of victimization, one would
expect that the people most likely to get murdered are also
the ones most likely to own a gun—But this is not always the
case: “While blacks are significantly more likely than whites
to be gun homicide victims, blacks are only about half as
likely as whites to have a firearm in their home (41% vs.
19%)” (Pew Research Center, 2014). Accordingly, whereas
young men are more likely to be murder victims, old men are
more likely to own a gun (General Social Survey, 2015).
One could argue that, although the objective risk of homi-
cide victimization is low, the fact there is any risk at all suf-
fices to justify owning a gun and even carrying it on a daily
basis. This raises the second question—namely, whether
guns are reliably used in assault scenarios. It is easy to imag-
ine that this is the case when an assault comes from strangers
who attack them on the street or break into their house—the
prototypical self-defense scenarios described in a regular
column in the American Rifleman, the flagship magazine of
the National Rifle Association (NRA). These stories often
portray a heroic armed citizen who defends himself or her-
self against armed strangers, be it burglars or violent maniacs
(O’Neill, 2007).
Yet a study of homicides in the United States, between
1980 and 2008, found that strangers commit only a fifth of
all homicides (Siegel et al., 2014); four fifths are committed
by the victim’s family, friends, or acquaintances. Furthermore,
actual defensive gun use is rare: In ≤1% of reported incidents
of violent crime did the would-be victim exercise resistance
via firearm, according to the 1987-1990 National Crime
Victimization Survey (NCVS; McDowall & Wiersma, 1994;
Planty & Truman, 2013). Similar results were reported by
Hemenway and Solnick (2015): In only 127 of 14,145 (<1%)
crime incidents in which victims were present was a gun
used as a means of armed resistance. Hemenway and Solnick
(2015) concluded,
The NCVS data provide little evidence that self-defense gun use
reduces the likelihood of victim injury during crime. The data do
suggest that using a gun may be useful at preventing property
loss, but not more effective than protective action using other
weapons. (p. 27)
What about defense against terrorist attacks? Given that
mass shootings, such as the 2016 Orlando nightclub attack,
are homicides in which some stranger kills several people in
a single attack, one might think guns are effective for self-
defense in such situations. However, despite the fact that
mass shootings have increased in recent years (A. P. Cohen,
Azrael, & Miller, 2014), the chance of being involved in such
a shooting is extremely small (Bagalman, Caldwell, Finklea,
& McCallion, 2013). Furthermore, John P. Blair, the coau-
thor of a book Active Shooter Events and Response (Blair,
Nichols, Burns, & Curnutt, 2013), warns against drawing a
gun in an active shooter event, because the police—or other
concealed-carry gun owners—might mistake such individu-
als for the active shooter and kill them (Achenbach, 2016).
To summarize, there is no clear case that gun ownership
corresponds with objective risk of attack. The people at most
risk of homicide victimization are among the least likely to
Stroebe et al. 3
own guns, and guns rarely get used in the very assault sce-
narios for which they are intended. Yet the main reason that
people say they own guns is for protection/self-defense. This
suggests—in line with a long history of psychological theo-
rizing (e.g., Koffka, 1935; Lewin, 1917)—that it is people’s
perceived risk and not their objective risk of attack that
explains their need to own a gun for self-defense. It further
suggests that—in addition to the specific threat of victimiza-
tion—some diffuse or generalized threats might contribute to
people’s fears.
A Two-Component Theory of
Defensive Gun Ownership
The starting point for the present research is the evidence
that the need for protection/self-defense is indeed a main
motivator of gun ownership. Our two-component theory
refers to the antecedent threat perceptions that give rise to the
need, followed by the consequent utility judgments that
motivate specific types of gun ownership as well as beliefs
about how to use guns for self-defense. Our theory differs
from criminological “fear of crime” theories in two key ways
(e.g., Cao et al., 1997; DeJong, 1997; Kleck et al., 2011;
Williams & McGrath, 1976). The first key difference is that
our theory predicts that the need for protection/self-defense
is driven by more than just a specific threat—the perceived
risk of lifetime assault victimization—but also by a diffuse
threat emanating from the belief that the world is a danger-
ous and unpredictable place. This generalized belief is not
linked to any single objective threat in particular and has
been found to be associated with a conservative worldview
and prejudice against minorities (Altemeyer, 1988; see also
Duckitt, 2001). The second key difference is that our theory
predicts that the same perceived threats that motivate people
to purchase a handgun also shape their beliefs about defen-
sive gun use and the fundamentality of their Second
Amendment rights. This belief system includes not just the
right to shoot to kill in a threatening situation but also oppo-
sition to laws that could interfere with their gun rights.
Antecedents of a Need for Protection/
Self-Defense: A Threat-Construal Theory
A direct test of the fear of crime hypothesis would relate gun
ownership to perceived risk of victimization. Although crim-
inological studies have found mixed support for the fear of
crime interpretation, Kleck and colleagues (2011) blame the
lack of support on weaknesses in the research, such as failure
to find out whether a respondent is actually the owner of a
gun (rather than being a household member) and whether the
gun is actually a handgun (rather than some other type of
gun). When Kleck et al. assessed the association between
personal ownership of a handgun for protection and self-
defense, they found a significantly positive association with
perceived risk of crime. This suggests that, in line with the
fear of crime explanation (Cao et al., 1997; DeJong, 1997;
Kleck et al., 2011; Williams & McGrath, 1976), the Perceived
Lifetime Risk of Assault (PLRA)—the specific threat that
one might at some point become a victim of a crime—is one
of the factors that contribute to a person’s need for protec-
tion/self-defense. Yet if this were the only reason, one would
assume gun ownership to be much more reliably associated
with homicide rates, with homicide being the life-ending
crime. This has not been the case: Gun ownership has con-
tinuously decreased from 49% in 1960 to 42% in 2014
(Gallup, 2014) even though there had been fluctuations in
homicides over the same time period—including a steep
increase in homicide rates between 1960 and 1992 followed
by 49% decline from 9.3 homicides in 1992 to 4.7 in 2011
(U.S. Department of Justice, 2013).
In a departure from previous theorizing, we therefore pos-
tulate that there also exists a nonspecific and diffuse threat
feeding the need for protection and self-defense—namely, the
belief that the world is a dangerous place. Altemeyer (1988)
argued that the “Belief in a Dangerous World” (BDW) reflects
a “worldview,” or a system of beliefs about the nature of the
social world, specifically about what people are like. Altemeyer
(1988) developed a scale measuring the “Belief in a Dangerous
World,” ranging from one extreme—the view that the world is
inherently dangerous, unpredictable, and threatening—to the
opposite extreme—that the world is a secure, stable, and basi-
cally safe place. A modified BDW scale was developed by
Duckitt (2001); example items include the following: “There
are many dangerous people in our society, who will attack
someone out of pure meanness, for no reason at all.” “Any day
now, chaos and anarchy could erupt around us. All signs are
pointing to it” (Duckitt, 2001, p. 69).
BDW is closely associated with right-wing authoritarian-
ism, with the idea that “High RWAs are scared. They see the
world as a dangerous place, as society teeters on the brink of
self-destruction from evil and violence” (Altemeyer, 1988, p.
52). BDW is also correlated with negative attitudes toward
minorities (r = .39; Duckitt, 2001), and data from the
American Election Study suggest symbolic racism correlates
with U.S. Whites having a gun in their homes (O’Brien,
Forrest, Lynott, & Daly, 2013), as well as with U.S. Whites’
opposition to gun control policies (Filindra & Kaplan, 2016;
O’Brien et al., 2013). Thus, the BDW may be a useful indica-
tor of a range of concerns that indirectly link to gun owner-
ship but are not necessarily focused on the fear/perceived
risk of attack.
Our threat-construal theory of the need for protection/self-
defense postulates that two independent construals of threat—
the diffuse BDW and the specific PLRA—independently
induce a need for protection/self-defense. The notion that
threat can be construed at different levels is consistent with
social-cognitive theories on how objects, ideas, or events can
be construed either in low-level and specific terms or in
abstract and generalized terms (e.g., Trope & Liberman,
4 Personality and Social Psychology Bulletin
2010; Wegner & Vallacher, 1986). One could perceive a spe-
cific threat of criminals and violent maniacs, or a generalized
threat of a dangerous world and a society at the brink of col-
lapse. Our main objective, in distinguishing levels of threat
construal, is to demonstrate how social-psychological theo-
ries may be uniquely equipped to show that threat perceptions
need not be grounded in objective reality to trigger a need for
self-defense—They need only make one feel threatened.
Consequences of a Need for Protection/
Self-Defense: Motivated Reasoning
With regard to connecting American gun ownership to a gen-
eral motivational theory, beliefs in diffuse and specific
threats undermine individuals’ need for physical security—
which is among the most basic human needs (e.g., Maslow,
1943). In a dangerous world where one’s own risk of victim-
ization is perceived to be high, the goal of preventing harm to
oneself and one’s family may loom large. Under these cir-
cumstances, the more individuals value the goal of protec-
tion/self-defense, the more they are likely to search for
means that are effective for attaining it (Kruglanski et al.,
2002). This search may include the acquisition of firearms as
well as a desire to maximize their instrumentality.
There are many ways to protect oneself against a possible
attack, such as attending a course on self-defense techniques or
buying pepper spray. People should only buy a gun if they per-
ceive it as an effective means for self-defense. Given that hand-
guns are generally perceived as the most utilitarian weapon for
self-defense (Cook & Ludwig, 1996; Kleck et al., 2011), the
need for protection/self-defense should mainly be associated
with handgun ownership. Long guns are instead purchased for
reasons such as hunting or target/sport shooting.
The type of motivated reasoning we propose is rooted fun-
damentally in a utility judgment—a desire to maximize the
instrumentality of their chosen means of self-defense.
Therefore, the need for self-defense is likely to promote not
just handgun ownership but also beliefs about how handguns
can and should be used by their owners—namely, that gun
owners have a right to shoot or kill other people in self-defense
and that a well-armed society is a safe society. Given that work
on motivated reasoning suggests that people’s needs and
desires affect their beliefs (Bélanger, Kruglanski, Chen, &
Orehek, 2014; Bélanger, Schori-Eyal, Pica, Kruglanski, &
Lafrieniere, 2015; Dunning, 1999; Kunda, 1990), we predict
that the strength of the need for self-defense could also shift
one’s reasoning to be more supportive of broad Second
Amendment rights—the element of the U.S. Constitution that
guarantees a right to bear arms. This could lead many gun
owners to take more adversarial positions on gun control laws
and oppose restrictive government policies with respect to
guns—the reason being that gun control laws may reduce indi-
viduals’ ability to avail themselves of a means perceived as
effective to attainment of the goal of personal
security. We further propose that the extent to which the need
for protection/self-defense predicts these downstream beliefs
about gun use and gun rights will be mediated by perceptions
that guns are indeed effective means of protection/self-defense.
If the need does not relate to perceptions that gun possession
has utility for protection/self-defense, the gun owner may not
be as motivated to advocate for Second Amendment rights.
These gun-related beliefs are of considerable interest
because federal officials (mostly Democrats) started to push
gun control legislation after the unconscionable 2012 mass
shooting at Sandy Hook Elementary in Newtown, Connecticut,
in which 20 young children and six of their caretakers were
murdered by a 20-year-old gunman (Crabtree, 2012; Stroebe,
2016). To date, no federal legislation has passed—and gun
rights advocates instead maintain that the best means of
defense is for more people to be armed and to scrap laws
imposing “gun free zones” in places. As the Executive Vice
President of the NRA—Wayne LaPierre—argued, after the
Sandy Hook shooting, the only thing that stops a bad guy with
a gun is a good guy with a gun (Memmott, 2012).1 In his opin-
ion, shared by many gun owners, it is the presence of guns that
deters crime and thus gun free zones only attract criminals. We
will test whether such advocacy ultimately stems from the pur-
suit of protection and self-defense and a desire to maximize
the perceived utility of guns.
Testing the Two-Component Theory of
Defensive Gun Ownership
In the empirical section of our article, we test our proposed
model using data from two surveys of male gun owners and
nonowners. Study 1 used the survey data to compare gun
owners with nonowners to tap possible differences in their
gun-related beliefs. Studies 2 and 3 focused exclusively on
the gun owner survey to test the predictions derived from our
two-component theory. Specifically, Study 2 focused on pre-
dictions about gun ownership, whereas Study 3 tested pre-
dictions about gun owners’ beliefs. Study 4 was a replication
of Studies 2 and 3 using new data. New surveys were con-
ducted because a horrific mass shooting happened in Orlando
just as we finished collecting data for the initial surveys. We
thus sought to assess the potential impact of the mass shoot-
ing on gun-related beliefs and evaluate the stability of our
model in a replication.
Study 1: Assessing Gun-Related Beliefs
of Male Gun Owners and Nonowners
Participants
In total, 839 men in the United States were recruited via the
market research firm Qualtrics Panels to complete the study
online (May 31-June 11, 2016). We focused on men because
they are much more likely to be gun owners than women
(Gallup, 2013). Participants were recruited on the basis of gun
Stroebe et al. 5
ownership (n = 404 gun owners, n = 435 nonowners) and also
by region of country, age, education, and income. Regions
included Midwest (n = 208), West (n = 118), Northeast
(n = 158), and South (n = 355); region did not differ by gun
ownership (F < 1). Median age category was “35-44,” educa-
tion was “some college,” and income was US$35,000 to
US$50,000 per year. Education did not differ by gun owner-
ship (F < 1.8), but gun owners tended to report slightly higher
age and income categories (Fage = 21.95, p < .001, ηp
2 = .026;
Fincome = 23.78, p < .001, ηp
2 = .028).
Among the gun owners, 82.2% (n = 332) owned a hand-
gun and 77.2% (n = 312) owned a long gun. Of the owners of
long guns, 82.4% (n = 257) owned a shotgun, 60.6%
(n = 189) owned a precision rifle, and 24.7% (n = 77) owned
a modern sporting rifle (AR-15 or AK-style, “MSR”). The
mean number of guns owned was 4.06 (SD = 4.37; range =
1-30); 25.1% owned one gun, 24.5% owned two guns, and
the rest (50.4%) owned three guns or more. Seventeen did
not report the number of guns owned.
Procedure
Participants first self-reported their demographics, above, to
screen them for gender (males only) and gun ownership (“Do
you own a gun?”), and to also ensure a wide range of demo-
graphics (based on region, age, education, and income). The
informed consent explicitly stated the purpose of this
research was to assess beliefs, attitudes, and experiences
regarding gun ownership and the use of firearms. To mini-
mize biased language or terminology, the questionnaires
were designed with feedback from two professionals in gun
sales and manufacturing. Participants then reported their
threat-related beliefs and gun ownership in counterbalanced
order.
BDW (diffuse threat). To assess participants’ belief in a dan-
gerous and threatening social world, we used the 10-item
version of Altemeyer’s (1988) BDW scale developed by
(Duckitt, 2001).2 Items were rated from 1= strongly disagree
to 7= strongly agree (α = .84).
PLRA (specific threat). To assess participants’ perception of
their lifetime risk of assault, they were asked, “What do you
estimate is the likelihood the following will happen in your
lifetime (in your future)?” There were three items: “You will
be mugged,” “You will be violently attacked,” and “Your
home will be invaded by an armed burglar” (rated 1= not likely
at all to 7 = very likely, α = .90).
Type of gun owned (defensive gun ownership). Gun owners were
asked about the different guns they owned, whereas nonown-
ers were asked which type of gun they would buy if they
would ever purchase a gun. Below the question were four
labeled images of guns representing our main categories of
interest: Handgun, Precision rifle, MSR (AR-15, AK-style),
Shotgun; they could also select: Any other class III/National
Firearms Act weapons and not applicable/none of the above.
Main reasons for owning a gun (e.g., need for protection/self-
defense; value rating). Gun owners reported the extent to
which each of the following was a reason they owned a gun:
Protection/Self-Defense, Hunting, Target/Sport Shooting,
Constitutional Right/Second Amendment, Collecting Guns/
Hobby, and Other. Participants gave their ratings on a scale
ranging from 1 (not important/not applicable) to 7 (very
important). The questions were reframed for nonowners to
be about the main reasons they might consider buying a
gun.3
Perceived effectiveness of gun possession (utility judgment). Both
gun owners and nonowners were asked, “How effective is
gun possession as a means of . . .” and were first presented
with our motivation of interest, “Protection and Self-Defense”
(rated 1 = not effective at all to 7 = extremely effective).
Justification to shoot an intruder. Next, we assessed various
beliefs about gun rights, starting with justification for when a
gun can and should be used. Participants were presented with a
vignette to examine in which type of situation they would feel
justified to shoot an intruder: “If a person encounters an
intruder, in his home, in the middle of the night, how justified
is it for him to . . .” and on separate screens, participants rated
the justification of five actions “. . .fire a warning shot to scare
off the intruder”; “. . . shoot and wound the intruder”; “. . . shoot
and kill the intruder”; “. . . shoot the intruder, even if the intruder
is already trying to flee the home”; and “. . . shoot the intruder,
even if the homeowner is otherwise alone and can get out
safely” (rated 1 = not at all justified to 7 = totally justified,
α = .84). The last two items represent a belief one does not
have a duty to avoid violence (retreat) if assaulted in one’s
home—a key feature of “Stand your Ground” and “Castle”
laws (also referred to as “Make my day law”).
Right to kill. The next set of questions focused directly on
situations in which participants felt that a man has the right
to kill another person: “A man has a right to kill another man
in a case of self-defense,” “A man has a right to kill a person
to defend his family,” and “A man has a right to kill a person
to defend his home” (rated 1 = strongly disagree to
7 = strongly agree, α = .74; Cohen & Nisbett, 1994).
Gun rights advocacy. Five items assessed support for a well-
armed citizenry free from government regulation. The first
three items assessed opposition to various forms of govern-
mental gun control that were under national debate at the
time this study was conducted: “In general, do you believe
the laws covering the sale of firearms should be made more
strict, less strict, or kept as they are now?” (rated 1 = much
less strict to 7 = much more strict, [reverse-coded]); “Do you
6 Personality and Social Psychology Bulletin
support or oppose some kind of registry of all guns, at least
at the state government level?” (rated 1 = strongly oppose a
gun registry to 7 = strongly support a gun registry, [reverse-
coded]); “Do you support or oppose laws that create ‘gun
free zones’ at schools and other public places?” (rated 1 =
strongly oppose “gun free zones” to 7 = strongly support
“gun free zones,” [reverse-coded]). The next two items
assessed a belief that mass gun ownership promotes order in
society, “In general, if more people had guns, there would be
less crime,” and an adversarial view of the current federal
government, “In general, the federal government wants to
take people’s guns away” (rated 1 = strongly disagree to 7 =
strongly agree, α = .70).
Experience with victimization. We included additional mea-
sures to provide discriminant validity for the specific and per-
ceived risk—namely, PLRA may be more associated with
actual incidents of victimization, whereas the diffuse risk
(BDW) may not be. Participants completed a series of ques-
tions related to their personal experience with guns and crime,
including whether they knew a specific instance in which
someone close to them was the victim of violent crime (coded
1 if they selected “family member,” “close friend,” “neigh-
bor,” “someone else close [text-entry option],” or coded 0 for
“no” or “not sure”). Each participant also answered whether
he had ever been a victim of a violent crime himself (coded 1
for “yes” and 0 for “no”).
Political orientation. Participants also reported their political
orientation (rated 1 = extremely conservative to 9 = extremely
liberal), and whether they identify with any particular politi-
cal party (coded 1 for Republican, Libertarian, or if they
gave a text entry of any other pro-Second Amendment party,
for example, Constitutionalist, and 0 for other types of party
identification, for example, Democrat, Socialist, No).
Preliminary Analyses—Effects of
Questionnaire Order
We had counterbalanced the gun ownership and perceived
threat questionnaires, and initial one-way ANOVAs indi-
cated small but significant main effects of questionnaire
order on both BDW, F(1, 837) = 5.23, p = .022, ηp
2 = .006,
and PLRA, F(1, 837) = 9.04, p = .003, ηp
2 = .011. Irrespective
of gun ownership, merely asking about gun ownership before
asking about threats slightly increased BDW (M = 4.38,
SD = 1.14 vs. M = 4.21, SD = 1.08) and PLRA (M = 3.60,
SD = 1.53 vs. M = 3.28, SD = 1.59). There were no main
effects of order on any of the other variables (Fs < 3.3,
ps ≥ .07). Apparently, simply asking questions about per-
sonal gun ownership increased subsequent perceptions of
specific as well as diffuse threat.
We also tested for interactions between questionnaire
order and gun ownership. Separate 2 × 2 ANOVAs on each
variable, in a 2 (gun owner vs. nonowner) × 2 (questionnaire
order) design, yielded only one interaction on the variable
Gun Rights Advocacy, F(1, 835) = 5.00, p = .026, ηp
2 = .006;
this interaction was above and beyond the main effect of gun
ownership, F(1, 835) = 53.76, p < .001, ηp
2 = .060. The inter-
action was driven by a slight widening of the main effect of
gun ownership on Gun Rights Advocacy when participants
were asked about gun ownership first (Mgun owner = 4.01,
SD = 1.28 vs. Mnonowner = 3.12, SD = 1.33), as opposed to
when asked about threat first (Mgun owner = 3.79, SD = 1.40 vs.
Mnonowner = 3.32, SD = 1.39).
Main Analysis and Discussion
Study 1 compared gun owners and nonowners on our survey
questions to assess the extent to which the two groups differ.
Table 1 presents the means of the responses of these two
groups and whether the differences are significant. Except
for political orientation (i.e., self-ratings on the conservative–-
liberal dimension) and firing a warning shot to scare off an
intruder, all differences are highly significant. Gun owners
perceive more threats than nonowners: They report higher
BDW and PLRA. They also report protection/self-defense as
a stronger reason for gun ownership than nonowners and
believe more strongly that gun possession is an effective
means of self-defense. They report higher justification to
shoot an intruder, higher right to kill, and they advocate
greater gun rights in society in general. Gun owners are also
more likely to identify with Republican, libertarian, or con-
stitutionalist parties.
A closer look at the data indicated it was specifically
handgun owners who reported these high levels of self-
defense motivation: A one-way ANOVA comparing handgun
owners to long gun only owners, as well as nonowners, sug-
gested a significant difference, F(2, 830) = 30.76, p < .001,
ηp
2 = .07. Handgun owners reported higher Reason:
Protection/Self-Defense, M(n=332) = 6.18 (95% confidence
interval [CI] = [5.98, 6.38]), than both the owners of long
guns only, M(n=68) = 4.84 (95% CI = 4.40, 5.28]), and the
nonowners, M(n=433) = 5.24 (95% CI = [5.06, 5.41]). The
exact same pattern was observed for Effective: Self-Defense,
ANOVA: F(2, 833) = 30.95, p < .001, ηp
2 = .07.
Handguns are the weapons of choice for defensive gun
ownership. But how can we explain these differences?
Studies 2 and 3 used the survey data to test whether a two-
component model of defensive gun use can explain both
handgun ownership and gun use in self-defense.
Study 2: Identifying the Motivational
Determinants of Gun Ownership
To test our two-component theory of defensive gun owner-
ship, Study 2 focuses exclusively on respondents who
owned a gun. To establish whether handguns were in fact
Stroebe et al. 7
bought mainly for protection/self-defense, survey respon-
dents were asked to indicate the main reasons for owning a
gun. Although most gun owners owned several guns (M =
4.06, SD = 4.37), we had asked for the main reasons they
owned a gun and were offered five alternatives (Protection/
Self-Defense, Hunting, Target/Sport Shooting,
Constitutional Right/Second Amendment, and Collecting
Guns/Hobby). We expected handgun owners in particular
to report higher Protection/Self-Defense motivation. A
comparison of the small group of gun owners, who either
owned only a handgun or only a long gun, supports this
assumption: Fully 45.6% of “long gun only” owners gave
their highest rating to something other than Reason:
Protection/Self-Defense, but only 12.4% of handgun own-
ers did the same. This pattern was also supported in a mul-
tiple regression analysis on all gun owners, showing that,
when simultaneously regressing ownership of all four
types of guns (handgun, shotgun, precision rifle, MSR;
ownership coded 1 = “yes” and 0 = “no”), over Protection/
Self-Defense, only handguns were positively associated
with defensive gun ownership, B = 1.25 [95% CI = 0.83,
1.67], t(398) = 5.88, p < .001. Thus, as expected, handgun
ownership was mainly associated with self-defense. This
was less the case for long gun ownership.
Main Analysis
We then tested the first part of our two-component theory of
defensive gun ownership. We conducted a path analysis with
structural equation modeling (SEM) using Mplus software
(Muthén & Muthén, 1998-2011). The covariance matrix for the
SEM was based on 399 gun owners (n = 5 had missing data);
bivariate correlations are in Table 2. We used maximum likeli-
hood parameter estimates with a Satorra–Bentler correction to
address nonnormality in the data (MLM scaling correction fac-
tor = 1.05). The model provided close approximate fit to the
data: comparative fit index (CFI) = 1.00/Tucker–Lewis index
(TLI) = 1.03; root mean square error of approximation
(RMSEA) = 0.00 (95% CI = [0.00, 0.05]); standardized root
mean square residual (SRMR) = 0.02; χ2(5) = 3.03, p = .70 (see
Hoyle, 1995; Kline, 2005).
Figure 1 shows the standardized path coefficients for our
model. In line with predictions, the paths of BDW and PLRA
to need for protection/self-defense (Reason: Protection/Self-
defense) were significant, with the link from BDW being
stronger than the link from PLRA. Notably, the indirect path
from BDW → Protection/Self-Defense → handgun owner-
ship was not only significant, b = 0.08, SE = .03, p = .002, but
also larger than the indirect path for PLRA, b = 0.04, SE =
Table 1. Mean Differences by Gun Ownership Status.
Variable
Gun owners
M (SD)
Nonowners
M (SD)F statistic ηp
2
Belief in a Dangerous World 4.43 (1.07) 4.16 (1.14) 11.73*** .01
Perceived Lifetime Risk of Assault 3.75 (1.58) 3.14 (1.50) 32.41*** .04
Reason for gun ownership
Protection/self-defense 5.96 (1.67) 5.24 (2.03) 31.29*** .04
Effectiveness of gun possession for
Protection/self-defense 6.12 (1.26) 5.42 (1.68) 46.40*** .05
Justified to shoot an intruder (five items) 5.12 (1.51) 4.49 (1.57) 34.89*** .04
. . . Fire a warning shot 6.07 (1.68) 5.96 (1.61) 1.08 (ns) .00
. . . Shoot and wound the intruder 5.82 (1.68) 5.34 (1.82) 15.98*** .02
. . . Shoot and kill the intruder 5.30 (1.95) 4.33 (2.18) 45.49*** .05
. . . Shoot even a fleeing intruder 3.71 (2.23) 3.06 (2.05) 19.21*** .02
. . . Shoot even if one can escape 4.71 (2.14) 3.78 (2.15) 39.63*** .05
Right to kill (three items) 6.07 (1.12) 5.32 (1.56) 63.19*** .07
A man has a right to kill another man in a case of self-defense 6.11 (1.29) 5.36 (1.76) 49.12*** .06
A man has a right to kill a person to defend his family 6.43 (1.05) 5.77 (1.60) 49.19*** .06
A man has a right to kill a person to defend his home 5.66 (1.66) 4.83 (1.93) 44.92*** .05
Gun rights advocacy (five items) 3.89 (1.35) 3.21 (1.36) 52.67*** .06
Agree more guns leads to less crime 4.59 (1.96) 3.46 (2.00) 67.68*** .08
Agree government wants to take guns 4.57 (2.09) 3.86 (2.12) 23.60*** .03
Support stricter laws on gun sales (r) 4.67 (1.54) 4.89 (1.59) 4.00* .01
Support state or federal gun registry (r) 4.47 (2.16) 5.23 (1.88) 29.65*** .03
Support presence of “gun free zones” (r) 4.55 (2.13) 5.14 (2.03) 16.75*** .02
Political orientation (→ liberal) 4.60 (2.25) 4.70 (2.16) 0.43 (ns) .00
Right Political Party ID (e.g., Republican, libertarian) 40.3% 27.1% Wald: 16.27*** Exp(B): 1.35
Note. Results were unchanged when controlling for demographic variables. (r) = reverse-coded item.
*p < .05. **p < .01. ***p < .001.
8 Personality and Social Psychology Bulletin
.02, p = .024. There was no reliable path from self-defense to
long gun ownership. Long gun ownership was instead pre-
dicted by Hunting. This supports the antecedent threat con-
strual part of our two-component theory of gun ownership.
Note that we also tried other models: Adding Target/Sport
Shooting did not predict handgun or long gun ownership
when Protection/Self-Defense and Hunting were already
included in the model. We also explored Effectiveness: Self-
Defense as a mediator in the gun owner model, but including
it worsened model fit: CFI = 0.94/TLI = 0.90; RMSEA =
0.07 (95% CI = [0.04, 0.10]); SRMR: 0.05; χ2(10) = 26.77,
p < .001; though, there were still indirect paths from handgun
ownership to BDW, b = 0.04, SE = .01, p = .008, and hand-
gun ownership to PLRA, b = 0.02, SE = .01, p = .044.
Discriminant Validity Analyses
The two types of threat (BDW and PLRA) were only moder-
ately correlated (r = .45, p < .001), but we nevertheless con-
ducted discriminant validity analyses to ensure that PLRA and
BDW indeed represent independent types of threat. We specifi-
cally assessed whether the better predictor of PLRA was a spe-
cific experience with victimization and whether the better
predictor of BDW was a general worldview. We thus regressed
BDW on the variables political orientation—rated on a 9-point
scale from conservative to liberal, and their personal experi-
ence with victimization—namely, whether they personally
know a victim of a violent crime or whether they had ever been
a victim of a violent crime (note that 59.2% of gun owners
reported knowing a victim and 22.8% reported having been a
victim). First, we regressed BDW on political orientation,
know a victim, and been a victim. Results indicated that politi-
cal orientation was the strongest predictor of BDW, B = −0.10
(95% CI = [−0.15, −0.06]), t(398) = −5.43, p < .001. Politically
conservative gun owners reported higher BDW. Although
know a victim and having been a victim also predicted BDW—
knowing a victim: B = 0.24 (95% CI = [0.02, 0.46]), t(398) =
2.11, p = .035; been a victim: B = 0.13 (95% CI = [0.002,
0.26]), t(398) = 1.99, p = .047—hierarchical regression indi-
cated know a victim and been a victim explained only a com-
bined 2.2% of variance in BDW, ΔF(2, 399) = 4.51, p = .012,
whereas political orientation explained a further 4.6%, ΔF(1,
399) = 19.60, p < .001. Next, we regressed PLRA on political
orientation, know a victim, and been a victim, and found only
the victimization items to be significant predictors—know a
victim: B = 0.52 (95% CI = [0.19, 0.85]), t(398) = 3.10, p =
.002; been a victim: B = 0.24 (95%
CI = [0.05, 0.43]), t(398) = 1.99, p = .015. In a hierarchical
Table 2. Correlations Between Type of Gun Owned and Reasons for Gun Ownership.
Variable 1 2 3 4 5
1. Belief in a Dangerous World (BDW) —
2. Perceived Lifetime Risk of Assault Perceived (PLRA) .45*** —
3. Reason: Protection/self-defense .32*** .25*** —
4. Reason: Hunting .07 .07 −.03 —
5. Handgun ownership .10* .13* .29*** −.07 —
6. Long gun ownership −.02 −.02 −.10* .42*** −.21***
*p < .05. **p < .01. ***p < .001.
Figure 1. Study 2: Path diagram for gun ownership model.
Note. Values are standardized coefficients (and standard errors in parentheses). Gray lines are covariances (automatically generated in Mplus).
*p < .05. **p < .01. ***p < .001.
Stroebe et al. 9
regression, the victimization items accounted for a combined
6% of the variance in PLRA, ΔF(2, 399) = 12.79, p < .001,
whereas political orientation did not predict a significant
amount of variance (ΔR2 = .005, ΔF = 2.07, p = .151). In line
with our assumptions, the better predictor of PLRA was a spe-
cific experience with victimization; in turn, the better predictor
of BDW was the person’s general political orientation.
Discussion
The reasons given for gun ownership were consistent with
expectations. Although some long gun owners also gave self-
defense as a main reason for owning a gun, only handguns
were positively associated with defensive gun ownership
when simultaneously regressing ownership of all four types of
guns over Protection/Self-Defense. That nearly all gun owners
own several guns is consistent with the fact that although there
are enough guns to arm every American, only 30% of
Americans actually report owning a gun (Gallup, 2014).
The path analysis supports our reasoning for the anteced-
ent threat-construal component of our two-component the-
ory. The need for protection/self-defense is driven by two
kinds of perceived threats: one diffuse and abstract (BDW),
the other specific and concrete (PLRA). The diffuse threat
derives from the belief that the world is a dangerous and
unstable place, populated by bad people, and that society is
at the brink of collapse. The specific threat is reflected by
people’s estimate of their risk of becoming the victim of
some kind of violent assault during their lifetime. We also
observed that BDW and PLRA are independent: PLRA was
more strongly associated with knowing or having been the
victim of violent crime, whereas BDW was more strongly
associated with political conservatism. It is worth noting that
experience with victimization explained only 6% of the vari-
ance in PLRA, which suggests that even perceived risk might
stem from sources other than direct personal experience.
Mass media are likely to have a powerful impact in shaping
people’s worldviews. Exposure to stories involving deceit,
crime, and violence might foster learning that the world is a
hostile, mean, and violent place (Saleem & Anderson, 2012).
There is also evidence that frequent exposure to television
news, particularly local news, is associated with an increased
perception of crime risks on both personal and societal levels
(Romer, Hall Jamieson, & Aday, 2003).
Our path analysis indicated BDW was the stronger pre-
dictor of need for protection/self-defense. This could make it
difficult to conduct persuasion campaigns aimed at dissuad-
ing handgun owners of the need to own a gun for self-
defense. There are strong arguments to be made that for most
people, guns are neither needed for self-defense nor very
useful when the need arises. If the need for protection/self-
defense had mainly been driven by specific risk perception,
persuasion campaigns could be aimed at reducing the per-
ceived risk. In contrast, the BDW is a broader system of
beliefs about the nature of the social world and what people
are like. Such worldviews are extremely difficult to influ-
ence because they are based on childhood socialization
(Altemeyer, 1988; Duckitt, 2001). Worldviews are also
coherent belief systems—so changing any one specific belief
would make it inconsistent with many other beliefs
(Kruglanski & Stroebe, 2005).
That the need for protection/self-defense predicts only
handgun ownership and not long gun ownership is to be
expected, but it nevertheless provides several advances for
understanding the motivational bases of gun ownership. First,
this possibility is often overlooked: As Kleck and colleagues
(2011) argued, the failure of many studies to observe a link
between gun ownership and fear of crime or perceived risk of
victimization could be due to such studies trying to connect
this fear to gun ownership generally and not ownership of a
handgun in particular. Second, our threat-construal theory
further suggests that fear of crime should not simply be
assessed by the specific threat of becoming the victim of vio-
lent crime but also by tapping the diffuse threat that the world
is a dangerous place. That said, it is worth noting that our
threat-construal theory may still not capture the full range of
perceived threats: The correlations between BDW and PLRA
and Protection/Self-Defense, though significant, are only
small to medium (see Table 3; J. Cohen, 1977). This suggests
that there must be other threats, not assessed in our study, that
make people feel they need to own a gun for self-defense.
Could protection/self-defense just be a pretense? We con-
sidered the possibility that men did not want to admit that the
real reason they bought a gun was that owning a gun made
them feel powerful. We did ask our respondents how effective
gun ownership was as a means of empowerment—but the
correlation of this variable with handgun ownership was not
significant: r(n=404) = .032, p = .520. Subsequent research
should use more subtle measures to probe more thoroughly
the relation between the sense of empowerment and gun
ownership.
Study 3: Need for Protection/Self-
Defense and Motivated Reasoning
Next we tested how the need for protection/self-defense
could extend beyond handgun ownership to affect beliefs
about how they can and should be used. Based on data from
the original survey of gun owners, we examined whether the
strength of handgun owners’ need for protection/self-
defense motivates the type of reasoning that justifies shoot-
ing and killing another person and for having broad rights to
acquire and carry guns. More specifically, given that hand-
gun ownership is itself only a means to an end, we consider
whether gun rights advocacy is itself motivated by a utility
judgment regarding the effectiveness of guns for satisfying
one’s goal of security. Handgun owners have chosen gun
ownership as a means of attaining security, but they may
nevertheless differ in perceptions of utility—that is, whether
they believe that guns are actually effective means of
10 Personality and Social Psychology Bulletin
protection/self-defense. Based on classic expectancy-value
theory (e.g., Feather, 1982), values and expectations com-
bine to predict behavioral tendencies. We thus tested whether
the relationship between need for protection/self-defense
and gun rights beliefs is mediated by the extent to which gun
owners perceive guns as effective means of protection/self-
defense. If the need does not relate to perceptions of utility,
the person might only nominally own guns for protection/
self-defense and may not be especially motivated to advo-
cate for broad rights to carry and use them.
Given that the need for protection/self-defense pre-
dicted only handgun, but not long gun ownership, the pres-
ent study focused on the 330 handgun owners from the
survey (n = 2 had missing data). The study focused on
three dependent measures, namely, (a) the handgun own-
ers’ responses to the questions about how to deal with an
intruder, (b) the type of situation that would give a man the
right to kill another person, and (c) the individual’s advo-
cacy of societal rights to gun ownership and use. The mean
responses to the five intruder questions were used for the
“Justification to shoot” dependent measure. Similarly, the
mean responses to the three right to kill items constituted
the “Right to kill” dependent measure and the mean
responses to the five gun advocacy beliefs represented the
“Gun rights advocacy” measure. Bivariate correlations are
in Table 3.
Main Analysis and Discussion
The SEM path analysis used maximum likelihood parameter
estimates with a Satorra–Bentler correction (MLM scaling
correction factor = 1.12). The model provided reasonable fit to
the data: CFI = 0.96/TLI: 0.92; RMSEA: 0.06 (95% CI =
[0.03, 0.10]); SRMR: 0.06; χ2(11) = 21.15, p = .007. All indi-
rect paths—from BDW to Reason: Protection/Self-Defense, to
Effectiveness: Protection and Self-Defense, to the dependent
variables—were significant: “Justification to shoot,” b = 0.04,
SE = .01, p = .009; “Right to kill,” b = 0.05, SE = .02, p = .007;
and “Gun rights advocacy,” b = 0.04, SE = .01, p = .006. All
the indirect paths of PLRA were also significant: “Justification
to shoot,” b = 0.03, SE = .01, p = .032; “Right to kill,” b = 0.04,
SE = .02, p = .027; and “Gun rights advocacy,” b = 0.03,
SE = .01, p = .028.
Figure 2 presents the motivated reasoning model for
handgun owners. Consistent with our expectation based on
expectancy-value theory (e.g., Feather, 1982), the same vari-
ables that predicted handgun ownership also predict justifi-
cation to shoot and right to kill. In this model, effectiveness
of self-defense mediated the link between need for protec-
tion/self-defense and the dependent variable. It is possible
that the instrumentality of handguns, as a means for self-
defense, becomes more salient when gun owners are asked
about their gun use in self-defense situations. In line with our
Table 3. Correlations Between Handgun Owners’ Beliefs About Gun Use and Ownership.
Variable 1 2 3 4 5 6
1. Belief in a Dangerous World (BDW) —
2. Perceived Lifetime Risk of Assault (PLRA) .43*** —
3. Reason: Protection/self-defense .27*** .26*** —
4. Effectiveness: Protection and self-defense .25*** .21** .53*** —
5. Justify shoot .13* .13* .28*** .34*** —
6. Right to kill .20*** .15** .27*** .42*** .63*** —
7. Gun rights advocacy .13* −.02 .10 .33*** .18*** .29***
*p < .05. **p < .01. ***p < .001.
Figure 2. Study 3: Path diagram for the motivated reasoning model.
Note. Values are standardized coefficients (and standard errors in parentheses). Gray lines are covariances automatically generated in Mplus.
*p < .05. **p < .01. ***p < .001.
Stroebe et al. 11
predictions, handgun owners, who perceived a greater threat
(BDW, PLRA), not only felt a greater need for protection and
self-defense but also saw guns as a more effective means of
self-defense. The perceived effectiveness of guns as a means
for self-defense was positively associated with justification
to shoot intruders, the perceived right to kill in self-defense
or defense of home and family, and gun rights advocacy. All
the indirect paths were significant, from both BDW and
PLRA to the various dependent measures. Thus, the factors
that motivate individuals to buy a handgun are also posi-
tively associated with more extreme behavior in dealing with
intruders and more extreme beliefs in the justification for
killing another person.
Study 4: Assessing the Impact of
the Orlando Mass Shooting on Gun
Owners’ Beliefs
Studies 1 to 3 outline our two-component theory of defensive
gun ownership; in Study 4, we test our theory with new data
in the context of a major mass shooting. The survey data for
the initial studies were collected from May 30 to June 11,
2016. There was no data collection on June 12—the date of
the horrific mass shooting at the Pulse nightclub in Orlando,
Florida. As reported by the New York Times on June 12,
It was the worst act of terrorism on American soil since Sept. 11,
2001, and the deadliest attack on a gay target in the nation’s
history . . . The toll of 50 dead is larger than the number of
murders in Orlando over the previous three years. Of an
estimated 320 people in the club, nearly one-third were shot.
The casualties far exceeded those in the 2007 shooting at
Virginia Tech, where 32 people were killed, and the 2012
shooting at an elementary school in Newtown, Conn., where 26
people died. (Alvarez & Pérez-Pena, 2016)
Given that there is usually a spike in background checks for
gun sales after mass shootings (e.g., Crockett, 2016;
Rojanasakul & Migliozzi, 2016), we wanted to assess
whether the Orlando mass shooting was linked to an increase
in PLRA or BDW.
We therefore repeated our survey with an independent
sample of gun owners. Our aims were to examine the
impact of the Orlando mass shooting on gun owners’ belief
systems and to test the stability of our two-component the-
ory of gun ownership by replicating our models in a new
sample. The post-Orlando dataset was collected in the days
immediately following the Orlando mass shooting (June
13-22). For this sample, the same market research firm
recruited 495 male gun owners, in the United States, with
no resampling of participants from the pre-Orlando sample.
The pre- and post-Orlando samples of gun owners did not
differ significantly by demographics (region, age, educa-
tion, or income, Fs < 1.2).4
Main Analyses
To test the impact of the news of the Orlando mass shooting
on any of the threat and gun-related beliefs assessed in our
surveys, we first compared the mean responses on the mea-
sures of these beliefs between the samples of gun owners
assessed pre-Orlando and post-Orlando. With the exception
of PLRA, post-Orlando gun owners did not differ on any of
the belief dimensions. There were no significant differences
for reason and effectiveness of guns for self-defense, justifi-
cation to shoot and intruder, right to kill, gun rights beliefs,
rates of handgun and long gun ownership, or BDW (Fs <
1.85). The negligible tendency to report lower average PLRA,
F(1, 896) = 4.42, p = .036, ηp
2 = .005, is probably due to the
fact that we had added a fourth item, the lifetime risk of
“being present during a mass shooting” (M = 2.70, SD = 1.67)
to the post-Orlando survey.
We then applied the SEMs from Studies 2 and 3 to the
post-Orlando sample. Fit statistics were quite similar between
the pre- and post-Orlando datasets: When the gun ownership
model, from Study 2 (pre-Orlando), was applied to the post-
Orlando data, it again provided close approximate fit to the
data: CFI = 0.98/TLI = 0.96; RMSEA = 0.04 (95%
CI = [0.00, 0.08]); SRMR = 0.03; χ2(5) = 8.12, p = .15; scal-
ing correction factor = 1.32. Accordingly, when the motivated
reasoning model, from Study 3 (pre-Orlando), was applied to
the data of the handgun owners from the post-Orlando sam-
ple, it also provided reasonable fit to the data: CFI = 0.96/TLI
= 0.93; RMSEA = 0.05 (95% CI = [0.02, 0.08]); SRMR: 0.06;
χ2(11) = 23.42, p = .015; scaling correction factor = 1.23.
As illustrated in Figure 3, the most striking feature of the
post-Orlando data is their similarity to the pre-Orlando data:
The paths for the motivated reasoning model were basically
unchanged; the paths for the ownership model suggested
only slight strengthening between handgun ownership and
Hunting (from b = −0.06 to b = −0.15) and between Lifetime
Risk of Assault and Protection/Self-Defense (from b = 0.12 to
b = 0.20). We also replicated the observation, from the pre-
Orlando gun ownership model, that when we try to include
Effectiveness: Self-Defense as a mediator in the post-Orlando
gun ownership model, it only reduced model fit: CFI = 0.91/
TLI = 0.83; RMSEA = 0.08 (95% CI= [0.06, 0.11]); SRMR:
0.06; χ2(5) = 41.71, p < .001. Altogether, in the days immedi-
ately following the mass shooting, gun owners showed no
meaningful changes in defensive gun ownership tendencies,
and our two-component model remained stable for both gun
ownership and motivated reasoning.
Discussion
Although our post-Orlando survey was conducted in the
aftermath of the deadliest mass shooting by a lone gunman to
date, defensive gun ownership or gun use beliefs were virtu-
ally unaffected by it. The mass shooting did not affect gun
owners’ beliefs in the effectiveness of guns for
12 Personality and Social Psychology Bulletin
protection/self-defense, nor did it change their scores on the
BDW; it left hardly any impact on the PLRA and did not
appear to change their belief that there would be less crime if
more people had guns. Despite the apparent lack of any
direct effect of the Orlando mass shooting on defensive gun
ownership, this study achieved another important goal—
namely, to demonstrate that our models could be replicated
in a second independent sample. The fit of our models was
virtually unchanged pre- and post-Orlando.
General Discussion
Why do millions of Americans feel the need to own a gun to
defend themselves and their families? There does not appear
to be a great deal of evidence that gun ownership corre-
sponds with objective risk of victimization. To illustrate this
point, our article began with an interview with a citizen of
Wickenburg—one of the safest small towns in Arizona—
who carries a gun every day to defend himself and his fam-
ily. Given that Wickenburg had only one case of murder
between 2007 and 2012 (Criminal Records Database, 2016),
the citizen’s motivation for carrying a gun may not be rooted
purely in an objective risk of attack but also perceived risk.
In developing a psychological model of defensive gun own-
ership, we acknowledge that threat of crime—or more specifi-
cally the perceived risk of becoming the victim of violent
crime—is indeed one of the factors that motivate people to own
a handgun. Beyond this, however, there is also a diffuse threat,
emanating from the belief that the world is a dangerous place,
that feeds the need for protection/self-defense. Our survey data
suggest that these two construal levels of threat independently
induce a goal to attain security, which motivates individuals to
buy a handgun, but not a long gun, due to their utility as defen-
sive weapons. We further observed that, to the extent gun own-
ers perceive their guns as instrumental for self-defense, the
same factors would motivate gun owners to maximize the
instrumentality of their chosen means of self-defense through
advocating for broad freedoms to use and carry them.
We compared the gun-related belief systems of gun own-
ers to that of nonowners and observed it was specifically
handgun owners who perceived more threat than nonowners.
Handgun owners reported higher PLRA and higher BDW
than both nonowners and owners of long guns only; these
perceived threats ultimately motivate handgun owners to
maximize the instrumentality of their guns by advocating for
broader gun rights.
We tested our gun ownership and motivated reasoning
models with two independent samples of gun owners, one
interviewed before the Orlando mass shooting and one
immediately afterward. Model fit was good for both samples,
suggesting strong support for the validity of our two-compo-
nent theory. Altogether, we observed little impact of the
Orlando mass shooting on gun owners’ belief systems. This
could be a reflection of how belief systems anchored in basic
Figures 3. Post-Orlando (a) gun ownership and (b) motivated reasoning models.
Note. Values are standardized coefficients (and standard errors in parentheses). Gray lines are covariances automatically generated in Mplus.
*p < .05. **p < .01. ***p < .001.
Stroebe et al. 13
motivations (like the need for physical safety) are hard to
change, even in the face of new information. However, it
could also suggest gun owners had already accounted for the
risk of mass shootings before Orlando. Although mass shoot-
ings are increasing in frequency—and this may imply that
we live in dangerous times—it does not necessarily imply
that our personal risk of victimization has substantially
increased. Statistically, mass shootings are rare, and the like-
lihood to be present during such an attack is minimal. Less
than 1% of gun deaths are from mass shootings. In 2013,
there were 30 mass shootings, killing 137 victims (Welch &
Hoyer, 2013). In the same year, 505 people died due to acci-
dental discharge of a firearm, 11,208 due to gun homicides,
and 21,175 due to gun suicides (GunPolicy.org, 2016).
Finally, consider that our research was conducted in the
specific context of American culture wherein beliefs in the
primary role of guns for protection are deeply entrenched
and enshrined in the U.S. Constitution. It is possible that in
other cultural contexts wherein guns occupy a less promi-
nent place, beliefs that the world is dangerous and that
one’s personal risk of assault is high would not translate
into gun ownership and gun supportive beliefs. Even within
the United States, people may differ in their exposure to
media violence that shapes perceptions of both specific and
general threats as well as perceptions that guns are an effec-
tive and appropriate means of defense against such threats.
To summarize, the world may be a dangerous place, but
based on our review of studies, there is reason to doubt that
gun ownership can be predicted purely by an individual’s
objective risk of assault. Our data suggest handgun ownership
and advocacy is, at least in part, a psychological phenome-
non. Gun ownership is predicted by various levels of per-
ceived risk—including a diffuse BDW, which is not tied to
any single objective risk in particular. To the best of our
knowledge, this study is the first to show how social-cogni-
tive theories—in this case, different levels of threat construal
and matters of perceived utility—offer useful insight into
what motivates Americans to own handguns and advocate for
broad rights to carry and use them.
Authors’ Note
All authors contributed equally to this manuscript.
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, author-
ship, and/or publication of this article.
Notes
1. This quote is a motto among many Second Amendment advocates.
2. The gun industry experts advising us on the study recommended
we avoid use of the term “anarchy,” in Item 2 of the “Belief in
a Dangerous World” (BDW) scale, so we replaced it with “law-
lessness.” They also recommended use of the term “Modern
Sporting Rifle” (MSR) in place of “assault rifle.”
3. Participants then completed a cognitive task lasting about 5 min,
wherein we manipulated its difficulty and subsequently mea-
sured state affect (anxious, hostile, and quiescent affect). This
manipulation had main effects on the affect measures in the
expected directions (Fs > 23.28, ps < .001), but no main effects
on our variables of interest (Fs < 3.6, ps > .05).
4. In all, n = 45 completed the original survey and n = 450 com-
pleted a new version of the survey with extra questions about the
mass shooting. The second survey also included one additional
screening question, when assessing demographics at the begin-
ning, designed to exclude any potential participants who were
unaware the Orlando mass shooting had even occurred.
Supplemental Material
The supplemental material is available on the PSPB website.
References
Achenbach, J. (2016, June 17). How to improve your chances to
survive a mass shooting like Orlando. The Washington Post.
Retrieved from https://www.washingtonpost.com/news/post-
nation/wp/2016/06/17/how-to-survive-an-active-shooting-
like-orlando/?utm_term=.f24ea9dfdec9
Altemeyer, B. (1988). Enemies of freedom: Understanding right-
wing authoritarianism. San Francisco, CA: Jossey-Bass.
Alvareze, L., & Pérez-Pena, R. (2016, June 12). Orlando gun-
man attacks gay night-club, leaving 50 dead. The New York
Times. Retrieved from http://www.nytimes.com/2016/06/13/
us/orlando-nightclub-shooting.html?_r=0
Bagalman, E., Caldwell, S. W., Finklea, K. M., & McCallion, G.
(2013). Public mass shootings in the United States: Selected
implications for federal public health and safety policy
(Congressional Research Service, R43004). Retrieved from
https://journalistsresource.org/wp-content/uploads/2013/03/
MassShootings_CongResServ.pdf
Bélanger, J. J., Kruglanski, A. W., Chen, X., & Orehek, E. (2014).
Bending perception to desire: Effects of task demands, moti-
vation, and cognitive resources. Motivation and Emotion, 38,
802-814. doi:10.1007/s11031-014-9436-z
Bélanger, J. J., Schori-Eyal, N., Pica, G., Kruglanski, A. W., &
Lafrieniere, M.-A. (2015). The “more is less” effect in equifi-
nal structures: Alternative means reduce the intensity and qual-
ity of motivation. Journal of Experimental Social Psychology,
60, 93-102. doi:10.1016/j.jesp.2015.05.005
Blair, J. P., Nichols, T., Burns, D., & Curnutt, J. R. (2013). Active
shooter events and response. Boca Raton, FL: CBC Press.
Cao, L., Cullen, F. T., & Link, B. G. (1997). The social determi-
nants of gun ownership: Self-protection in an urban environ-
ment. Criminology, 35, 625-650.
Cohen, A. P., Azrael, D., & Miller, M. (2014, October 15). Rate of
mass shootings has tripled since 2011, Harvard research shows.
Mother Jones. Retrieved from http://www.motherjones.com/
politics/2014/10/mass-shootingsincreasing-harvard-research
Cohen, D., & Nisbett, R.E. (1994). Self-protection and the culture
of honor: The role of institutions in perpetuating norms of vio-
lence. Personality and Social Psychology Bulletin, 23, 1188-
1199. doi: 10.1177/0146167294205012
14 Personality and Social Psychology Bulletin
Cohen, J. (1977). Statistical power analysis for the behavioral sci-
ences. New York, NY: Academic Press.
Cook, P. J., & Ludwig, J. (1996). Guns in America. Washington,
DC: Police Foundation.
Crabtree, S. (2012, December 19). Biden to head task force on gun
violence causes. The Washington Times. Retrieved from http://
www.washingtontimes.com/news/2012/dec/19/biden-to-head-
task-force-on-gun-violence-causes/
Criminal Records Database. (2016). Wickenburg, Arizona criminal
records. Retrieved from https://www.instantcheckmate.com/
criminal-records/Arizona/Wickenburg/
Crockett, Z. (2016). What happens after mass shooting?
American buy more guns. Retrieved from http://www.vox.
com/2016/6/15/11936494/after-mass-shooting-americans-
buy-more-gun
DeJong, C. (1997). Differential indicators of defensive weapon
ownership: A comparison by gender. Journal of Criminal
Justice, 25, 517-525.
Duckitt, J. (2001). A dual-process cognitive-motivational theory
of ideology and prejudice. Advances in Experimental Social
Psychology, 33, 41-114.
Dunning, D. (1999). A newer look: Motivated social cognition and
the schematic representation of social concepts. Psychological
Inquiry, 10, 1-11.
Feather, N. T. (Ed.). (1982). Expectations and actions: Expectancy-
value models in psychology. Hillsdale, NJ: Lawrence Erlbaum.
Federal Bureau of Investigation. (2014). Uniform crime report:
Crime in the United States. Retrieved from https://ucr.fbi.gov/
crime-in-the-u.s/2014/crime-in-the-u.s.-2014/offenses-known-
to-law-enforcement/expanded-offense
Filindra, A., & Kaplan, N. J. (2016). Racial resentment and white’s
gun policy preferences in contemporary America. Political
Behavior, 38, 255-275. doi: 10.1007/s11109-015-9326-4
Gallup. (2013). Men, married, Southerners most likely to be gun
owners. Retrieved from http://www.gallup.com/poll/160223/
men-married-southerners-likely-gun-owners.aspx
Gallup. (2014). Guns: Historical trends. Retrieved from http://
www.gallup.com/poll/1645/guns.aspx
General Social Survey. (2015). Trends in gun ownership in the
United States, 1972-2014 (Final report). Retrieved from http://
www.norc.org/PDFs/GSS%20Reports/GSS_Trends%20
in%20Gun%20Ownership_US_1972-2014.pdf
GunPolicy.org. (2016). United States—Gun fact, figures and the
law. Retrieved from http://www.gunpolicy.org/firearms/region/
united-states
Hemenway, D., & Solnick, S. J. (2015). The epidemiology of self-
defense gun use: Evidence from the National Crime Victimization
Surveys 2007–2011. Preventive Medicine, 29, 22-27.
Hoyle, R. H. (Ed.). (1995). Structural equation modeling: Concepts,
issues, and applications. Thousand Oaks, CA: SAGE.
Hwang, K., & Murphy, K. (2014, August 20). Armed, out and
about: Evolving public carry laws expand gun rights. NBC
News. Retrieved from http://www.nbcnews.com/news/investi-
gations/armed-out-about-evolving-public-carry-laws-expand-
gun-rights-n184296
Kleck, G., Kovandzik, T., Saber, M., & Hauser, W. (2011). The effect
of perceived risk of victimization on plans to purchase a gun for
self-protection. Journal of Criminal Justice, 39, 312-319.
Kline, R. B. (2005). Principles and practice of structural equation
modeling. New York, NY: Guilford.
Koffka, K. (1935). Principles of gestalt psychology. London,
England: Routledge.
Koffka, K (1936). London: Routledge and Kegan Paul.
Kruglanski, A. W., Shah, J. Y., Fishbach, A., Friedman, R. S., Chun, W.
Y., Sleeth-Keppler, D., & Zanna, M. P. (2002). A theory of goal
systems. In M. P. Zanna (Ed.), Advances in experimental social psy-
chology (Vol. 34, pp. 331-378). San Diego, CA: Academic Press.
Kruglanski, A. W. & Stroebe, W. (2005). Attitudes, goals and
beliefs: issues of structure, function and dynamics. In D.
Aberracin, B. Johnson, & M. P. Zanna (Eds.), Handbook of
attitude research (pp. 323-368). Mahwah, NJ: Erlbaum.
Kunda, Z. (1990). The case for motivated reasoning. Psychological
Bulletin, 108, 480-498.
Lewin, K. (1917). Die Kriegslandschaft [The war landscape].
Zeitschrift für Angewandte Psychologie, 12, 440-447.
Maslow, A. H. (1943). A theory of human motivation. Psychological
Review, 50, 370-396. doi:10.1037/h0054346
McDowall, D., & Wiersema, B. (1994). The incidence of
defensive firearm use by US crime victims, 1987 through
1990. American Journal of Public Health, 84, 1982-1984.
doi:10.2105/AJPH.84.12.1982
Memmott, M. (2012, December 21). “Only a good guy with a
gun” can stop school shootings, says NRA. Breaking News
From NPR. Retrieved from http://www.npr.org/blogs/thetwo-
way/2012/12/21/167785169/live-blog-nra-news-conference
Muthén, L. K., & Muthén, B. O. (1998-2011). Mplus user’s guide
(6th ed.). Los Angeles, CA: Author.
O’Brien, K., Forrest, W., Lynott, D., & Daly, M. (2013). Racism,
gun ownership and gun control: Biased attitudes in US whites
may influence policy decisions. PLoS ONE, 8, e77552.
O’Neill, K. L. (2007). Armed citizens and the stories they tell. Men
and Masculinities, 9, 457-475. doi:10.1177/1097184X05281390
Pew Research Center. (2014). The demographics and politics of
gun-owning households. Retrieved from http://www.pewre-
search.org/fact-tank/2014/07/15/the-demographics-and-poli-
tics-of-gun-owning-households/
Planty, M., & Truman, J. L. (2013). Firearm violence, 1993–2011
(Special report, NCJ 241730). U.S. Department of Justice.
Retrieved from http://www.bjs.gov/content/pub/pdf/fv9311.pdf
Rojanasakul, M., & Migliozzi, B. (2016). After Orlando, gun sales
surged. Bloomberg News. Retrieved from https://www.bloom-
berg.com/graphics/2016-gun-sales/
Romer, D., Hall Jamieson, K., & Aday, S. (2003). Television news
and the cultivation of fear of crime. Journal of Communication,
53, 88-104.
Saleem, M., & Anderson, C. A. (2012). The good, the bad, and
the ugly of electronic media. In J. Dvoskin, J. L. Skeem,
R. W. Novaco, & K. S. Douglas (Eds.), Using social science to
reduce violent offending (pp. 83-101). New York, NY: Oxford
University Press.
Siegel, M., Negussie, Y., Venture, S., Pleskunas, J., Ross, C. S., & King,
C. (2014). The relationship between gun ownership and stranger
and nonstranger firearm homicide rates in the United States, 1981–
2010. American Journal of Public Health, 104, 1912-1919.
Stroebe, W. (2013). Firearm possession and violent death: A critical
review. Aggression and Violent Behavior, 18, 209-721.
Stroebe, W. (2016). Firearm availability and violent death:
The need for a culture change in attitudes toward guns.
Analyses of Social Issues and Policy, 16, 7-35. doi: 10.1111/
asap.12100
Stroebe et al. 15
Trope, Y., & Liberman, N. (2010). Construal-level theory
of psychological distance. Psychological Review, 117,
440-463.
US Department of Justice (2011). Homicide rate falls to lowest rate
in four decades. Retrieved from https://www.bjs.gov
ValuePenguin. (2015). Safest places in Arizona. Retrieved from
http://www.valuepenguin.com/2015-safest-places-arizona
Wegner, D. M., & Vallacher, R. R. (1986). Action identification. In
R. M. Sorrentino & E. T. Higgins (Eds.), Handbook of motiva-
tion and cognition: Foundations of social behavior (pp. 550-
582). New York, NY: Guilford.
Welch, W. M., & Hoyer, M. (2013, December 15). 30 mass
killings, 137 victims: A typical year. USA Today. Retrieved
from http://www.usatoday.com/story/news/nation/2013/12/15/
mass-killings-main/3821897/
Williams, J. S., & McGrath, J. H., III (1976). Why people own guns.
Journal of Communication, 26, 22-30. doi: 10.1111/j.1460-
2466.1976.tb01931.x
