ONE journal article from the selected five attached to the assignment
. (Attached to assignment when you click on the assignment link.)
3) Using the attached file
(Analyzing a Journal Article),
complete each section of the worksheet based on your chosen journal article. Do not copy and paste straight from the journal article.
Analyzing a Journal Article
Step 1: What is the objective of this study?
a. Write down the exact statement in which the authors describe
what they were testing. This may be provided as a purpose statement
or a hypothesis. Include quotation marks around the exact wording
and indicate page number(s).
b. Describe the purpose of the study, as you understand it, in your
c. What is the gap in the previous research that the authors are trying
to fill by conducting this study?
Step 2: What are the major findings of the study?
a. What are the authorsâ€™ major conclusions or findings? Include
quotation marks if using exact wording, and indicate page number(s).
b. Write those conclusions, as you understand them, in your own
Step 3: How did the authors test their hypothesis?
a. Briefly summarize the main steps or measurements that the
authors used in their methods. May include type of study (survey,
observation, etc.), how, where and when data was gathered, study
b. How did the authors analyze their data? What test(s) did they use?
Step 4: How reliable are the results?
a. Do the authors suggest any problems or limitations with their
methodology? Do you see any problems or limitations with their
b. Do the conclusions made by the author make sense to you? Are the
conclusions too broad or too narrow based on what was actually done
in the study?
Step 5: What is the importance of this research study?
a. Write (in your own words) the significant contributions of the
research in this article as reported by the authors.
b. What are the implications of the study for the way we think about
Step 6: How does this study relate to other research in this field?
a. What prior research do the authors discuss? Look at the reference
list. What are the topic/themes of the resources cited?
b. What are the next steps or future directions for this research as
described by the authors?
c. What questions about this topic do you have after reading this
â€¢ Student Worksheet: Analyzing a Journal Article. https://www.lib.uoguelph.ca/sites/default/files/student_worksheet.pdf. University of Guelph. 2013
â€¢ Information Literacy Toolkit: How to Read a Scholarly Journal Article. https://legacy.lib.utexas.edu/signaturecourses/resources/how-read-scholarlyarticle. University of Texas at Austin, University of Texas Libraries. 2013.
See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/348189803
Moral Panic, Fear of Crime, and School Shootings: Does
Article in Sociological Inquiry Â· May 2021
6 authors, including:
State University of New York at Oswego
34 PUBLICATIONS 429 CITATIONS
88 PUBLICATIONS 1,326 CITATIONS
Some of the authors of this publication are also working on these related projects:
Public Sociology Project View project
All content following this page was uploaded by Jaclyn Schildkraut on 07 January 2022.
The user has requested enhancement of the downloaded file.
Moral Panic, Fear of Crime, and School Shootings: Does
H. Jaymi Elsass, Texas State University
Jaclyn Schildkraut, State University of New York at Oswego
Ross Haenï¬‚er, Grinnell College
Brian V. Klocke, Independent Consultant
Eric Madï¬s, University of Washington Tacoma
Glenn W. Muschert, Khalifa University of Science and Technology
The concept of moral panic has been used to describe how society reacts to different threats, either real or perceived. Although most studies in this area rely on qualitative, historical data, recent efforts have employed quantitative applications of an
attributional model to understand attitudes consistent with different characteristics of
moral panic. The present study utilizes this model along with survey data from college
students across the ï¬ve U.S. regions to examine how perceptions of school shootings as
a moral panic vary by location. The ï¬ndings reveal signiï¬cant differences by location,
with students in the Southwest and West less likely than other regions to express hostile
attitudes about the shootings or believe they are happening at disproportional rates.
Southwestern students also were most likely to express concern related to defense (gun
rights), whereas individuals in the Northeast were least likely to. Broader considerations
for these and other ï¬ndings also are discussed.
Mass shootings, particularly those that occur in schools, continue to be a
cause for public concern, especially when punctuated by incidents that receive
considerable media attention, like the 2018 attack at Marjory Stoneman Douglas High School in Parkland, FL. With each breaking news story, fear and
apprehension over school shootings in America are renewed. Parents fear sending their children to school, and students worry that a similar attack could take
place at their educational institution (Graf 2018). Such violence, however, is
not limited to primary and secondary schools. Institutions of higher education,
including the University of Texas at Austin (1966), Virginia Tech (2007),
Northern Illinois University (2008), and, more recently, the University of North
Carolina Charlotte (2019), also have experienced such attacks.
The manner and frequency in which school shootings are presented by
and through the media have made these events appear almost epidemic (Fox
Sociological Inquiry, Vol. 91, No. 2, May 2021, 426â€“454
Â© 2021 Alpha Kappa Delta: The International Sociology Honor Society
MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME
and DeLateur 2014; Muschert 2007a; Muschert and Ragnedda 2010; Newman
2006). In reality, schools remain among the safest places for children (Fox and
Friedel 2018) and events like school shootings are statistically rare in the context of the broader U.S. crime picture (Schildkraut, Formica, and Malatras
2018b; Schildkraut and Elsass 2016). Despite this contrary evidence, however,
the fear and subsequent responses generated in the wake of mass shootings in
schools have led some researchers (Burns and Crawford 1999; Madï¬s 2016;
Springhall 1999) to liken the reactions surrounding this phenomenon to a moral
A term coined by Young (1971) and Cohen (1972), moral panic occurs
when responses to a perceived threat are disproportionate to its actual existence
(Hall et al. 1978). In the context of school shootings, such responses have
included increased security in schools that has created a nearly $3 billion per
year product market and routine discourse on hot-button issues including gun
control, mental health, and violent media (see, generally, Schildkraut, Elsass,
and Muschert 2016; Schildkraut and Muschert 2013, 2019). Beyond school
shootings, other perceived social threatsâ€”including juvenile deviance and
crime (Cohen 1972; Hay 1995; Welch, Price, and Yankey 2002), gangs
(McCorkle and Miethe 1998; Zatz 1987), drugs (Hier 2002), and terrorism
(Rothe and Muzzatti 2004)â€”also have been considered through the lens of
In examining these and other issues, researchers typically rely on historical, qualitative data to better understand their impacts on society. Subsequently,
Goode and Ben-Yehuda (1994a, 1994b) introduced an attributional modelâ€”
emphasizing concern, hostility, consensus, disproportionality, and volatilityâ€”
that has become a framework for other scholars in this area (see also Critcher
2008; Klocke and Muschert 2010; Krinsky 2013). Schildkraut, Elsass, and Stafford (2015) extended this body of research by quantitatively applying Goode
and Ben-Yehudaâ€™s model to examine college studentsâ€™ perceptions of school
shootings. At the same time, their study allowed for examination of individuallevel responses to a macro-level collective phenomenon (Schildkraut et al.
Still, despite such advances, additional opportunities to expand this
research exist. Newman and Hartman (2019), for example, have called for spatial examinations of the impact of mass and school shootings on public attitudes. To date, however, virtually no research has considered the relationship
between location and perceptions about these events, either broadly or within
the context of moral panic theory. Accordingly, the present study seeks to
determine whether such a relationship exists. In this paper, we utilize data collected at ï¬ve universities from different regions of the United States. Respondents were asked about their beliefs related to school shootings that were
H. JAYMI ELSASS ET AL.
consistent with the ï¬ve attributes of Goode and Ben-Yehudaâ€™s (1994a, 1994b)
model, as well as questions related to their fear of crime and perceived risk of
victimization (see also Schildkraut et al. 2015). Based upon previous research
(e.g., Althaus, Cizmar, and Gimpel 2009; Dor e, Ort, Braverman, and Ochsner
2015; Lemiuex 2014), we hypothesized that perceptions of school shootings as
a moral panic would differ based on the location of the respondents.
Review of the Literature
According to Cohen (1972), who typically is credited with the development of the concept, moral panic can be described as:
A condition, episode, person or group of persons emerges to become deï¬ned as a threat to
societal values and interests; its nature is presented in a stylized and stereotypical fashion by
the mass media; the moral barricades are manned by editors, bishops, politicians, and other
right-thinking people; socially accredited experts pronounce their diagnoses and solutions;
ways of coping are evolved or (more often) resorted to; the condition then disappears, submerges, or deteriorates. (p. 9).
The corresponding responses offered by ofï¬cials, including politicians and
pundits, are what actually are indicative of a moral panic and typically are out
of proportion with the actual threat (Hall et al. 1978). Such responses typically
are harsher when the person, group, or event is perceived to be a greater threat
to society (Hall et al. 1978). In other words, â€œobjective molehills have been
made into subjective mountainsâ€ (Jones, McFalls, and Gallagher 1989:341),
and it is these contexts from which perceived threats, through the lens of a
moral panic, acquire their meaning (Krinsky 2013).
To supplement Cohenâ€™s (1972) processual model, Goode and Ben-Yehuda
(1994a, 1994b) highlighted ï¬ve attributes to provide a comprehensive framework through which to understand the characteristics typical of many moral
panics (see also Critcher 2008; Klocke and Muschert 2010; Krinsky 2013). The
ï¬rst of these attributes is concern, which is indicative of a heightened or intensiï¬ed anxiety level over a perceived social threat (Goode and Ben-Yehuda
1994a, 1994b). The manifestation of an â€œus-versus-themâ€ mentality is exemplary of hostility that accompanies a moral panic (Burns and Crawford 1999;
Cohen 1972; Goode and Ben-Yehuda 1994a, 1994b). Consensus occurs when
there is broad agreement about the existence of a threat and its potential to
cause harm; when public concern of the threat exceeds its actual occurrence,
the panic has become disproportional (Goode and Ben-Yehuda 1994a, 1994b;
see also Miller 2013). Finally, volatility occurs when the panic, which suddenly
erupts, disappears as abruptly (Goode and Ben-Yehuda 1994a, 1994b); this also
MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME
may be assessed based on how long a perceived threat remains part of the public discourse (see, generally, Downs 1972; McCombs and Zhu 1995).
School Shootings as a Moral Panic
Following the shooting at Columbine High School, Burns and Crawford
(1999) applied Goode and Ben-Yehudaâ€™s attributional model to highlight the
attackâ€™s contribution to the national panic over school shootings (see also
Springhall 1999). Such examination was particularly relevant as Columbine is
considered by many to be a watershed moment (Schildkraut and Muschert
2019). At the same time, the amount of attention that Columbine received in
the media and by the public, both in the immediate aftermath and even
20 years after the attack, cemented it as a â€œcultural legacy,â€ which is a form of
a moral panic that can be sustained over time (Goode and Ben-Yehuda
1994a:158). Not only did Columbine go on to become the archetypal model to
which all other school shootings that followed it would be compared (Kalish
and Kimmel 2010; Larkin 2007, 2009; Muschert 2007b; Muschert and Larkin
2007), it also became a cornerstone in the discourse about youth violence in
America (Muschert 2002; Schildkraut and Muschert 2019). Therefore, understanding how Columbine and subsequent events are situated within the attributes of moral panic can lend a greater understanding of the panic itself.
Concern. According to Goode and Ben-Yehuda (1994a, 1994b), concern
related to a perceived threat must be able to be measured in concrete and
explicit ways. The most accessible way to do so is through media reports
(Burns and Crawford 1999), particularly given the amount of coverage that
school shootings generate and the fact that the media serve as the main source
of information for approximately 95 percent of the general population (Graber
1980; Surette 2015). Hundreds of stories were aired about Columbine in the
ï¬rst month after the shooting on both cable news network programs (Muschert
2002) and evening news broadcasts (Robinson 2011; see also Maguire,
Weatherby, and Mathers 2002); the shooting was the biggest news story of
1999 and the third most followed of the decade after the Rodney King verdict
(1992) and the crash of TWA ï¬‚ight 800 (1996; Pew Research Center for the
People and the Press 1999). Comparable airtime was devoted to covering the
Virginia Tech shooting, also the biggest news story of its week, nearly eight
years later (Pew Research Center for the People and the Press 2007).
Additionally, more than 10,000 news stories were published about Columbine
in the top 50 national newspapers in 1999 (Newman 2006), with 170 stories
appearing in just The New York Times in the ï¬rst 30 days after the shooting
(Chyi and McCombs 2004). Similarly, 130 articles appeared in The Times in
H. JAYMI ELSASS ET AL.
the month following the 2012 shooting at Sandy Hook Elementary School
(Schildkraut 2014, 2016; Schildkraut and Muschert 2014).
Concern also may be measured by the responses to an event, such as legislative activity (Burns and Crawford 1999). In relation to school shootings,
ï¬rearms are a necessary component (compared to other weapons that would
make it an attack rather than a shooting), and thus, the majority of efforts in
the aftermath of these events typically focus on restricting ï¬rearms (Schildkraut
and Carr 2020). In the year after Columbine, for example, more than 800 bills
were introduced at the state and federal level (Soraghan 2000), including a
number of pieces designed to extend background checks (e.g., universal checks,
including gun shows) and eliminate straw purchases (Schildkraut and Hernandez 2014). Attempts to regulate assault-style weapons also have come in the
wake of school shootings (Schildkraut and Carr 2020). Despite the fervor of
activity, however, few bills are signed into law (Schildkraut and Carr 2020;
Schildkraut and Hernandez 2014), which also may serve as an indicator of the
imbalance between the saliency of the threat and the threat itself.
Hostility. Certain youth, either individually or as part of a collective
group, may be subject to increased hostility in the form of demonization,
criminalization, or even alienation when they exhibit problematic behaviors,
including violence (Burns and Crawford 1999). School shootings like
Columbine shifted perceptions away from the idea that violence was solely an
inner-city problem perpetrated by minority youth (Tonso 2009), highlighting
the fact that no community was or is immune to such tragedy (Frymer 2009;
Larkin 2007). As a result, students in educational institutions across the nation
who shared similar traits with the Columbine perpetrators and other school
shooters, including choices of clothing (e.g., trench coats) and music (e.g.,
â€œGoth rockâ€ like Marilyn Manson or Rammstein), were considered cause for
concern (Frymer 2009; Madï¬s 2020). Similarly, students who were not elites
or jocks but instead part of out-groups in high schools have also been regularly
proï¬led as the next potential school shooter (Frymer 2009; Larkin 2007; Tonso
Consensus. When there is agreement that a threat is â€œreal, serious, and
caused by the wrongdoing of group members and their behavior,â€ consensus is
said to have been achieved (Goode and Ben-Yehuda 1994a:157). Though such
agreement may not affect most people nor be universally accepted, it must be
widespread (Burns and Crawford 1999). Following Columbine, school
shootings became a signiï¬cant cause for national concern (Muschert 2009), with
the conversation about safety in such institutions transcending from Littleton to
all suburban high schools in the country (Altheide 2009b). Taken together, the
ampliï¬ed fear over school safety (Burns and Crawford 1999; Fox and DeLateur
MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME
2014; McCarthy 2015; Wike and Fraser 2009), coupled with a sense of disorder
and lack of control (Altheide 2009a), are markers of a moral panic.
Consensus among members of society about a problem can lead to a
desire to â€œdo somethingâ€ in order to remedy the damage caused by the issue
and restore â€œmoral orderâ€ (Burns and Crawford 1999:149). This reaction can
â€œinvolve strengthening the social control apparatus of the society, including
tougher or renewed rules, increased public hostility and condemnation, more
laws, longer sentences, more police, more arrests, and more prison cellsâ€
(Goode and Ben-Yehuda 1994b:30) and often is reï¬‚ected in public opinion
polls following mass shootings (see, e.g., Saad 1999). Responses to school
shootings have run the gamut, from hardening schools (e.g., Crawford and
Burns 2015) to implementing zero-tolerance policies for any behavioral infraction (Kupchik 2010). The typical public discourse following such attacks
focuses on condemning the â€œusual suspectsâ€ of guns, mental health, and violent
media (Schildkraut et al. 2016; Schildkraut and Muschert 2013, 2019), and
steps are taken, albeit often unsuccessfully as noted, to legislate the issues that
are perceived to have been the catalyst for the shooting (Schildkraut and Carr
2020; Schildkraut and Hernandez 2014; Soraghan 2000; Vasilogambros 2018).
Disproportionality. When attempting to understand how a moral panic
originates, it is imperative to compare the amount of attention the problem
garners compared to its actual occurrence (Burns and Crawford 1999; Muschert
and Ragnedda 2010). Disproportionality, a key to understanding the existence
of such panic, occurs when the perceived threat exceeds the magnitude of the
actual threat (Waddington 1986). The over-exaggeration of the scale of the
issue may be promoted by the media or may stem from overly severe ofï¬cial
reactions (Goode and Ben-Yehuda 1994a; Waddington 1986). In the context of
school shootings, this problem is further exacerbated by a lack of a precise
deï¬nition and dataset to understand the true breadth of the problem
(Schildkraut and Elsass 2016). Despite the perception that these events are
reaching epidemic-like proportions, schools, as noted, are among the safest
places for children to be (Fox and Friedel 2018). In fact, less than 2 percent of
youth homicides occur at educational institutions (Cornell 2015; Holland et al.
2019). Also contributing to the disproportional understanding about school
shootings is the fact that, despite their sensational yet newsworthy nature, not
all attacks receive equitable attention in and by the media; the emphasis on the
most extreme (and lethal) examples further distorts the perceptions about such
events (Schildkraut, Elsass, and Meredith 2018a).
Volatility. The concept of volatility, which Burns and Crawford (1999)
actually omit from their examination of Columbine, refers to the sudden
eruption and abrupt diminution of social threats (Goode and Ben-Yehuda
H. JAYMI ELSASS ET AL.
1994a). In other words, events like mass shootings are unpredictable in the
context of where and when they will occur. Consequently, measuring the
concept of volatility is particularly difï¬cult; one proposed solution to overcome
this challenge is to consider the length of time the phenomenon receives public
attention. In the context of media coverage, Downs (1972) refers to this as the
â€œissue-attention cycle,â€ in which stories have a certain shelf life before being
replaced by other topics of interest to satisfy the publicâ€™s need for new content.
Issues like money and politics enjoy longer shelf lives, averaging 3.5â€“4 years
(McCombs and Zhu 1995). Coverage of school shootings, on the other hand,
typically only lasts one month (at most) before being replaced by other news
headlines (Chyi and McCombs 2004; Schildkraut and Muschert 2014). In other
words, school shootingsâ€”in the context of public and media attentionâ€”are
particularly volatile in nature because the attention to them is short-lived.
A Quantitative Application. As noted, nearly all applications of the moral
panic framework, including those incorporating Goode and Ben-Yehudaâ€™s
attributional model, have relied on qualitative, historical data. In an effort to
extend the body of research to overcome this limitation, Schildkraut, Elsass, and
Stafford (2015) previously quantitatively applied Goode and Ben-Yehudaâ€™s
framework to better understand what characteristics of college students predicted
attitudes consistent with moral panic over school shootings, both collectively and
across the ï¬ve individual attributes. Their ï¬ndings indicated in the context of the
concern attribute (along the defense dimension), females and non-gun owners
were less supportive of carrying ï¬rearms for protection on campus, whereas those
respondents residing off campus were more amenable to such a proposal
(Schildkraut et al. 2015).1 Similarly, non-gun owners, respondents living off
campus, and those reporting greater levels of fear of crime also expressed
attitudes consistent with hostility, conceptualized as punitive attitudes toward
school shooters. Females were signiï¬cantly less likely than males to say that
school shootings were unpredictable (volatility). Respondents with greater levels
of fear of personal crime were signiï¬cantly more likely to agree that school
shootings were happening more frequently than their actual occurrence
(disproportionality); these same respondents also were signiï¬cantly more likely
to subscribe to the idea of a moral panic over these events. Schildkraut, Elsass,
and Staffordâ€™s (2015) quantitative study examining moral panic at one university
led to the current study to further the analysis using other predictors, with a
speciï¬c emphasis on location.
Why Should Location Matter?
Although research has yet to consider variation in attitudes consistent with
a moral panic over school shootings based on region speciï¬cally, there are
MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME
other studies that set a precedent for why location should be considered. Relating to school shootings speciï¬cally, Dor e et al. (2015) found that spatial, as
well as temporal, proximity to an attack can impact emotional and psychological reactions. Examining Twitter posts following the Sandy Hook shooting,
they found that the more spatially distant users were from Newtown, they were
less likely to tweet about the attack and less likely to express intense emotional
thoughts (Dor e et al. 2015). Importantly, the ï¬ndings also indicated that while
users were less likely to express sadness in their tweets the more spatially distant they were from the shooting, they also were more likely to express
increased anger and anxiety about the attack (Dor e et al. 2015). This ï¬nding in
particular highlights what Warr (2000) refers to as â€œaltruistic fear,â€ or fear for
others, rather than â€œpersonal fear,â€ where individuals are concerned about their
own safety. Moreover, in the context of fear of crime, anxiety represents concern over future victimization (Warr 2000), which mirrors the broader societal
responses to school shootings that call for prevention of the next attack.
Other studies also have found regional variations in topics related to these
events. Lemieux (2014), for example, found variability within the correlation
between legislation and crime rates, such that states with more permissive gun
laws, particularly those in the South and Midwest, have higher rates of both
gun deaths generally and homicides committed with ï¬rearms speciï¬cally.
Mixed ï¬ndings, however, have emerged related to the relationship between gun
laws, location, and the occurrence of mass shootings. Lemieux (2014) found
that states with restrictive policies, such as those in the Northeast and West,
experience more mass shootings and higher victim counts, which follows a
common argument that the perpetrators of these attacks purposefully select
â€œsoft targets,â€ or those locations where they are less likely to encounter armed
resistance (e.g., Hesterman 2015). Conversely, Friedel (2020) found that such
attacks are more likely to occur in states with higher level of gun ownership.
Further, as most people experience mass shootings through the media
rather than directly, differences between markets in the content and quality of
news can inï¬‚uence audience exposure (Althaus et al. 2009). Media use has
been found to vary by region, with local television news found to be more popular in the Midwest compared to the West, cable news being favored in southern states compared to the West and Midwest, and network news popular in
the Northeast and Southeast (including Gulf Coast) regions (Althaus et al.
2009). Similarly, news consumers in the northern regions of the United States
are more likely to read the newspaper than their counterparts in the southern
portion of the country (Althaus et al. 2009). In sum, these ï¬ndings highlight
the importance of considering the potential for regional differences in perceptions of attitudes related to a moral panic over school shootings given the
potential disparities in varying cultural and other processes at work.
H. JAYMI ELSASS ET AL.
The present study is guided by a key research question that seeks to
expand the scope of Schildkraut, Elsass, and Staffordâ€™s (2015) initial examination. Speciï¬cally, we ask do attitudes indicative of moral panic over school
shootings vary by location (region)? In order to answer this question, we
hypothesize that there will be regional variation among participantsâ€™ attitudes
about the problem of school shootings.
In order to assess potential regional variations, the survey instrument was
disseminated at ï¬ve universities of varying sizes in different parts of the country during the Fall 2012 (Southwest) and Spring 2013 (Northeast, Southeast,
Midwest, and West) semesters following IRB approvals at each institution.
Paper-and-pencil surveys were administered to undergraduate students at each
university, and data collection was completed in the same semester that it
began. Target sample sizes for each university were calculated using a 95 percent conï¬dence level and 5 percent margin of error based upon the institutionâ€™s
population. Convenience sampling of undergraduate courses then was utilized
to collect the number of surveys to meet the target sample size.2 Potential
respondents were notiï¬ed that their participation was voluntary, and no incentives were provided. A total of 1,906 surveys were collected across the ï¬ve
sitesâ€”Southwest (Texas), Northeast (New York), Southeast (Mississippi), Midwest (Ohio), and West (Washington).
Table 1 presents the descriptive statistics for respondents by region, as
well as the total sample. In the full sample, females represented the larger share
of respondents (58.4%), as did whites (66.8%). Students ages 21 and under
accounted for 74.8 percent of respondents. The majority of respondents identiï¬ed themselves as non-gun owners (75.9%) and indicated that they lived off
Nineteen Likert-scale questions operationalizing Burns and Crawfordâ€™s
(1999) moral panic indicators as they relate to school shootings were included
in the survey, as outlined by Schildkraut, Elsass, and Stafford (2015).
Responses ranged on a ï¬ve-point scale, from strongly disagree (1) to strongly
agree (5). While Goode and Ben-Yehudaâ€™s (1994a, 1994b) model of moral
panic included ï¬ve attributes (consensus, disproportionality, hostility, volatility,
and concern), Schildkraut, Elsass, and Staffordâ€™s (2015) factor analysis suggested that six factors were better suited for the phenomenon of school shootings with questions loading moderately to high on their respective factors. This
(N = 417)
(N = 412)
(N = 400)
(N = 235)
(N = 1,906)
Note: Variable frequency percentages may not total to 100.0 percent due to rounding error or missing data.
21 and younger
22 and older
Does not own
(N = 442)
Table 1 Descriptive Statistics for Respondents by Region and Total Sample
MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME
H. JAYMI ELSASS ET AL.
was the result of the concern attribute being split into two dimensions: one
focused on defense (including questions centering gun rights) and the other
measuring prevention (including questions pertaining to respondent attitudes
toward gun control). A list of the attributes, with their respective questions and
descriptive statistics, is presented in Appendix 1.
Prior to any analyses being conducted, however, patterns of missing data
â€”which ranged from 0.4 to 1.2 percent among the 19 questionsâ€”were
assessed using Littleâ€™s Test of Missing Completely at Random (MCAR). The
results indicated that the data were missing completely at random
(v2 = 627.072, p = .454). Subsequently, listwise deletion was used to exclude
these cases from the dataset, which Allison (2014) recommends as an appropriate and less biased way to handle missing data in larger sample sizes.
Further diagnostics also were conducted to assess whether the December
14, 2012 shooting at Sandy Hook Elementary School, which occurred during
the data collection period, posed a potential threat to internal validity of these
variables. Subsequently, chi-square analyses were conducted for each of the six
dependent measures by time (0 = pre-Sandy Hook, 1 = post-Sandy Hook). The
analyses revealed that statistically signiï¬cant differences were present among
four of the attributesâ€”consensus (v2 = 20.383, p < .05), disproportionality (v2 = 39.943, p < .05), hostility (v2 = 79.254, p < .001), and concernâ€“defense (v2 = 121.811, p < .001)â€”as well as the full moral panic scale (v2 = 85.816, p < .01). Accordingly, as a proxy for time, when location was included in the multivariate models, the Southwest region served as the reference category as it was the only school that completed data collection prior to the attack and therefore could simultaneously serve as a control for the potential history effect of the shooting. The four remaining schools began and completed their data collection after Sandy Hook. Independent Variables The present study also provided the opportunity for continued exploration of potential demographic differences in perceptions of school shootings as a moral panic (Schildkraut et al. 2015). Accordingly, additional dummy-coded measures were created for each independent variable, except for those involving fear of crime and perceived risk of victimization, with one category omitted from each to serve as the reference group (1 represented the presence of the characteristic, and its absence was coded as 0). Females were coded as 1, with males serving as the reference category. Three individual dummy variablesâ€” black, Hispanic, and other (including respondents reporting they were Asian or biracial), were created to measure race and ethnicity; whites served as the reference category. Age was dichotomized into two categories: 21 and under and 22 and older,3 with the younger group serving as the reference category. Southwest (N = 442) Northeast (N = 417) Southeast (N = 412) Midwest (N = 400) West (N = 235) F Post hoc analyses (Scheff e) *p â‰¤ .05; **p â‰¤ .01; ***p â‰¤ .001. 10.77 (2.10) 10.74 (1.86) 10.84 (1.92) 11.01 (1.62) 11.08 (1.82) 2.128 6.15 (2.92) 4.17 (2.41) 4.88 (2.56) 4.73 (2.29) 5.36 (2.87) 34.045*** NE < SW, NE < SE, NE < MW, NE < W, SE < SW, MW < SW, W < SW Concern 7.94 (1.91) 7.92 (1.77) 8.15 (1.68) 8.36 (1.55) 8.46 (1.78) 6.672** NE < MW, NE < W, (Prevention) SW < MW, SW < W Hostility 13.33 (4.15) 15.17 (3.26) 14.81 (5.00) 15.51 (3.15) 13.35 (3.67) 28.765*** W < NE, SW < NE, W < SE, SW < SE, W < MW, SW < MW Disproportionality 18.61 (5.02) 19.78 (4.37) 20.19 (4.87) 19.77 (4.60) 17.91 (5.31) 11.662*** W < NE, SW < NE, W < SE, SW < SE, W < MW, SW < MW Volatility 6.53 (1.96) 6.50 (1.82) 6.52 (1.64) 6.37 (1.78) 6.84 (1.87) 2.616* MW < W MP Full Scale 63.47 (9.57) 64.48 (8.24) 65.43 (9.45) 65.78 (8.00) 63.01 (9.37) 5.393*** W < MW, SW < MW Consensus Concern (Defense) Attribute Table 2 Means and Standard Deviations (SD) of Moral Panic Attributes and Global Measure by Region with One-Way ANOVA Statistics and Post Hoc Analyses (N = 1,906) MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME 437 438 H. JAYMI ELSASS ET AL. Two additional demographic independent variables assessing ï¬rearm ownership and residential status of the responding were included in the analysis. First, respondents were asked whether they owned a handgun, riï¬‚e, or shotgun. Those reporting ownership of at least one ï¬rearm were coded as being a gun owner, which was used as the reference category. Further, respondents were asked whether they live on or off campus, as earlier empirical ï¬ndings regarding fear of crime and college students (see, e.g., Tewksbury and Mustaine 2003) indicate that people who are spatially more distant from a crime locale (in this case, the campus itself) also will be less fearful. For this variable, those respondents reporting that they live on campus served as the reference category. The ï¬nal independent variables relate to fear of crime and perceived risk of victimization, which serve as conceptually and empirically distinct concepts (Ferraro and Grange 1987), with fear proposed as a cause of the perceived risk (Ferraro 1995). Further, fear and perceived risk of crime are conceptually distinct from moral panic. While the latter is considered a macro-level response to a perceived social problem, both fear and risk, as related to crime, are more indicative of individual-level responses to that problem (see, e.g., Elsass, Schildkraut, and Stafford 2014). In this study, 10 crimes and public order violations were used to create two identical panels of survey itemsâ€”one for fear of crime and one for perceived risk (Ferraro 1995; Grange and Ferraro 1989). Respondents were asked about their fear or perceived likelihood of being approached by a beggar or panhandler on the street, being conned out of money, having someone break into their home while they were present and while they were away, being raped or sexually assaulted, being murdered, being attacked with a weapon, having their car stolen, being mugged, and having oneâ€™s property vandalized (Ferraro 1995; Grange and Ferraro 1989).4 Responses for these panels ranged from 1 (not at all likely/afraid) to 10 (very likely/afraid). In order to differentiate between personal and property crime, additive scales were created. Personal crime included questions about having oneâ€™s home broken into while present, being murdered, and being attacked with a weapon (Ferraro 1995). In line with Schildkraut, Elsass, and Staffordâ€™s (2015) original research, rape was omitted in the present study as it proved to be highly collinear with fear and perceived risk of personal crime. This is to be expected if fear of rape drives fear of overall crime, albeit only for women (Warr 1984, 2000). Each scale related to personal crime was found to have acceptable internal consistency (Cronbachâ€™s a: fear = .886; risk = .830; see Tavakol and Dennick 2011). The remaining questions were added together to create the property crime measure, also with acceptable internal consistency among the respective questions included (Cronbachâ€™s a: fear = .879; risk = .785). MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME 439 Analysis and Findings Regional Variability in Moral Panic and Related Attribution The studyâ€™s research question centered on determining whether subscription to moral panic over school shootings varied by region. In order to answer this question and test the hypothesis that differences would be present based on respondentsâ€™ location, analysis of variance (ANOVA) was used with each of the individual attributes by region, as well as the complete moral panic scale. Scheff eâ€™s post hoc analysis also was employed to determine, when applicable, which groups differed from one another. The results of these analyses are presented in Table 2. As the ï¬ndings illustrate, for ï¬ve of the six attributes, excluding consensus, statistically signiï¬cant differences exist across regions. In ï¬rst examining the defense dimension of concern, respondents in the Northeast expressed less agreement that concealed handgun license (CHL) holders should be able to carry ï¬rearms on campus as compared to any of the other four regions (F = 34.045, p â‰¤ .001); conversely, more agreement was found among respondents in the Southwest compared to each of the other locations. For the opposite dimension of concern, prevention, which reï¬‚ects attitudes akin to gun control, respondents in both the Northeast and Southwest expressed less agreement as compared to those in either the Midwest or West (F = 6.672, p â‰¤ .001). Moreover, compared to individuals in the Northeast, Southeast, and Midwest, respondents in both the Southwest and West expressed more punitive attitudes (Hostility: F = 28.765, p â‰¤ .001) and stronger beliefs that school shootings are more likely to occur than they actually are (Disproportionality: F = 11.662, p â‰¤ .001). When examining volatility, differences in attitudes only were present between two regions: respondents in the Midwest were less likely than those in the West to believe that school shootings are unpredictable (F = 2.616, p â‰¤ .05). Finally, when considering the full scale, respondents in the Midwest were more likely than their counterparts in the Southwest and West to report greater moral panic over school shootings (F = 5.393, p â‰¤ .001). Multivariate Modeling Assessing Predictors of Moral Panic Supplemental analyses were conducted using ordinary least squares (OLS) regression to assess predictors of attitudes consistent with moral panic over school shootings, and the respondentsâ€™ location (region) served as the primary explanatory measures along with demographic variables as controls (Table 3). For both dimensions of concern (defense and prevention), hostility, and volatility, the signiï¬cant effects of region on these attributes were consistent with the ï¬ndings of the ANOVAs reported earlier. Notably, however, signiï¬cant effects 440 H. JAYMI ELSASS ET AL. related to the disproportionality attribute reversed compared to the prior analysis. Previously, the Scheff eâ€™s post hoc analysis of the ANOVA revealed that respondents in the Southwest group, which serves here as the reference category in the OLS regression, expressed less agreement about moral panic than individuals in the Northeast, Southeast, and Midwest. As the ï¬ndings of the regression indicate, however, these regions no longer are signiï¬cant when entered into the model with the other independent variables. Instead, respondents in the West were less likely than those in the Southwest to agree that school shootings are happening more frequently than they are (B = .081, p â‰¤ .01), whereas no signiï¬cant differences between the two regions related to the disproportionality attribute exist when examined independent of other potential predictors. Additional signiï¬cant correlates of attitudes consistent with a school shootings moral panic also were found. Among demographic correlates, in ï¬rst considering sex differences, when inspecting the effects of being female on each of the individual attributes (e.g., consensus), four signiï¬cant effects were found. The regression results indicate that being female was negatively correlated with volatility and the defense dimension of the concern attribute, but positively correlated with disproportionality and the prevention dimension of the concern attribute. Females, as compared to males, were less likely to say that licensed students and faculty should be able to carry guns on campus (B = .089, p â‰¤ .001). Females also were less likely than males to believe that school shootings are unpredictable events (B = .118, p â‰¤ .001). Conversely, females were more likely than their male counterparts to express concern over the prevention of school shootings (B = .062, p â‰¤ .05). Females also were more likely to believe that school shootings are more common events than they actually are (B = .125, p â‰¤ .001). Being female, however, was not found to be signiï¬cantly correlated with consensus or hostility. Race/ethnicity also was found to have a signiï¬cant correlation with four of the individual attributes of moral panic. With respect to consensus, blacks displayed signiï¬cantly less agreement than their white counterparts that the threat of school shootings are an imminent threat to the social order (B = .066, p â‰¤ .01). Concerning the two dimensions of the concern attribute, blacks were less likely than whites to support gun control (B = .065, p â‰¤ .01) as assessed through the prevention dimension, and also less likely to believe that CHL holders should be permitted to bring ï¬rearms on campus (B = .132, p â‰¤ .001), per the defense dimension. Additionally, as compared to whites, blacks offered less support for volatility (B = .015, p â‰¤ .001). Respondents identifying as races other than white, black, and Hispanic were less likely to express punitive attitudes related to school shootings as compared to their white Region Northeast .014 (.166) Southeast .050 (.154) Midwest .066 (.159) West .054 (.171) Gender Female .011 (.102) Age 22 and .043 (.120) older Race/Ethnicity Black .066 (.150)** Hispanic .004 (.158) Other .013 (.172) Gun Ownership Non-owner .004 (.124) Residence Off campus .010 (.114) Fear/Risk of Crime Fear .026 (.008) (Personal) Consensus .000 (.152) .051 (.141) .076 (.146)* .100 (.157)*** .062 (.094)* .034 (.110) .065 (.138)** .053 (.145) .045 (.157) .123 (.113)*** .005 (.104) .014 (.007) .089 (.130)*** .004 (.153) .132 (.192)*** .020 (.202) .037 (.218) .363 (.158)*** .006 (.145) .029 (.010) Concern (Prevention) .205 (.210)*** .101 (.196)*** .099 (.202)*** .062 (.219)* Concern (Defense) .118 (.014)** .030 (.209) .216 (.228)*** .020 (.279) .017 (.292) .052 (.315)* .032 (.221) .026 (.188) .143 (.304)*** .057 (.285)* .185 (.294)*** .026 (.316) Hostility .084 (.018)* .029 (.266) .053 (.289)* .022 (.353) .011 (.370) .029 (.402) .025 (.281) .125 (.239)*** .034 (.387) .044 (.362) .044 (.374) .081 (.401)** Disproportionality Table 3 OLS Regression Results for Moral Panic Attributes .028 (.008) .017 (.109) .030 (.118) .015 (.145)*** .095 (.152) .039 (.164) .012 (.115) .118 (.098)*** .049 (.158) .074 (.148)* .048 (.153) .086 (.164)** Volatility MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME 441 .031 (.012) .030 (.010) .246 .240 .007 (.021) .085 (.016)* .026 .017 .040 (.009) .047 (.007) Concern (Defense) .051 .043 .062 (.009) .066 (.015) .082 (.006) Concern (Prevention) .160 .153 .051 (.018) .015 (.030) .079 (.013) Hostility .178 .171 .082 (.023)* .054 (.038) .165 (.016)*** Disproportionality Note: Results presented as the standardized coefï¬cients with the standard errors in parentheses. *p â‰¤ .05; **p â‰¤ .01; *** p â‰¤ .001. Fear (Property) Risk (Personal) Risk (Property) R2 Adjusted R2 Consensus Table 3 (continued) .032 .023 .034 (.009) .068 (.016) .077 (.006) Volatility 442 H. JAYMI ELSASS ET AL. MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME 443 counterparts (B = .052, p â‰¤ .05). Hispanic respondents did not differ from whites in their attitudes across any of the moral panic attributes. Fear and perceived risk of both personal and property crime also were included in the models and yielded several signiï¬cant results. Speciï¬cally, with regard to hostility, respondents who reported greater levels of fear of personal victimization express more punitive attitudes about school shootings (B = .118, p â‰¤ .01). Similarly, concerning the disproportionality attribute, the greater the level of fear of personal victimization, the more likely the respondent was to agree that school shootings happen more frequently than they actually do (B = .084, p â‰¤ .05). Fear of personal crime, however, was not found to be signiï¬cantly correlated with any of the remaining attributes of moral panic. Perceived risk was correlated only with consensus: individuals who reported greater perceptions of their risk of personal victimization were less likely to express agreement about school shootings as a threat (B = .085, p â‰¤ .05). Moreover, the results indicated signiï¬cant differences between gun owners and non-gun owners in attitudes consistent with a moral panic over school shootings. As the results indicate, statistically signiï¬cant differences were found across four of the six attributes: both dimensions of concern, hostility, and disproportionality. Speciï¬cally, as it relates to concern, non-owners were less likely to support CHL holders carrying weapons on campus (Defense: B = .363, p â‰¤ .001) and more likely to support restrictions for possession (Prevention: B = .123, p â‰¤ .001) than gun owners. Additionally, non-gun owners also were more likely to express punitive attitudes about school shootings (Hostility: B = .216, p â‰¤ .001) and believe that they were happening more often than their actual occurrence (Disproportionality: B = .053, p â‰¤ .05). Finally, consideration is given to the potential for a broader moral panic about school shootings, measured by a general scale comprising all 19 questions (see also Schildkraut et al. 2015). Table 4 presents the results of the OLS regression. While the ANOVA results indicated that, for the general scale, respondents in the Midwest expressed attitudes more favorable to moral panic than individuals in the West and Southwest, only the latter held when the other variables were entered into the model (B = .107, p â‰¤ .01). Similarly, unlike in the OLS models for the individual attributes, females and non-gun owners did not differ signiï¬cantly in their attitudes about a general moral panic from males and gun owners, respectively. Blacks, however, were signiï¬cantly less likely to subscribe to a moral panic over school shootings as compared to white respondents (B = .053, p â‰¤ .05). Moreover, greater fear and perceived risk of personal crime were not correlated with such subscription. These collective ï¬ndings are discussed further in the next section. 444 H. JAYMI ELSASS ET AL. Discussion Research on moral panic has continued to evolve since the concept was introduced in the 1970s and, since Columbine, attention has shifted to understanding school shootings through this lens. Both Burns and Crawford (1999) and Springhall (1999) examined public response to Columbine as a function of moral panic using a qualitative approach. More recently, Schildkraut, Elsass, Table 4 OLS Regression Results for General Moral Panic Scale Moral panics Region Northeast Southeast Midwest West Gender Female Age 22 and older Race/Ethnicity Black Hispanic Other Gun Ownership Non-owner Residence Off campus Fear/Risk of Crime Fear (Personal) Fear (Property) Risk (Personal) Risk (Property) R2 Adjusted R2 .030 (.747) .051 (.698) .107 (.721)** .032 (.778) .040 (.461) .009 (.545) .053 (.681)* .021 (.715) .053 (.778) .026 (.561) .004 (.513) .076 (.035) .153 (.031)*** .001 (.073) .094 (.044)* .099 .091 Note: Results presented as the standardized coefï¬cients with the standard errors in parentheses. *p â‰¤ .05; **p â‰¤ .01; ***p â‰¤ .001. MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME 445 and Stafford (2015) considered perceptions related to school shootings as a phenomenon more broadly using a quantitative approach to test Goode and Ben-Yehudaâ€™s (1994a, 1994b) attributional theory. The present study extends this line of inquiry by speciï¬cally considering how such perceptions may vary based on location (see also Newman and Hartman 2019). When considering attitudes about moral panic and school shootings as a function of location, a number of interesting ï¬ndings emerged across the different attributes, with signiï¬cant differences across ï¬ve of six as well as the full scale. Interestingly, attitudes along the consensus attributeâ€”representing agreement of the existence of the threatâ€”are not signiï¬cant. In other words, the absence of signiï¬cant differences by region can be interpreted in and of itself to be consensus across all locations. Thus, regardless of where the respondent resides, there is collective agreement that school shootings are a problem, which similarly mirrors Schildkraut, Elsass, and Staffordâ€™s (2015) ï¬ndings of a lack of signiï¬cant predictors along the consensus attribute. Signiï¬cant differences across regions were found related to the defense dimension of the concern attribute, which represents attitudes consistent with gun rights. Speciï¬cally, respondents in the Southwest were more supportive of concealed carry policies for the purpose of defense against school shootings than individuals in any other region. This ï¬nding is not entirely unexpected, given Texasâ€™s pervasive and storied gun culture (see, e.g., Dizard, Muth, and Andrews 1999). In fact, although the policy was not in effect at the time the present survey was administered, Texas currently is one of just 10 states that allows individuals to carry ï¬rearms concealed on campus with a license, a measure introduced by some states after a number of shootings at colleges and universities, including Virginia Tech (National Conference of State Legislatures 2019). Conversely, respondents in the Northeast were less supportive of such policies compared to every other region. Notably, at the time the survey was administered, the State of New York recently had adopted one of the nationâ€™s most comprehensive gun control packages, the New York Secure Ammunition and Firearms Enforcement (SAFE) Act, several months earlier in response to the Sandy Hook Elementary School shooting. Although concealed carry on campus had never been permitted, the SAFE Act implemented other gun control measures, including universal background checks and enhanced penalties for illegal gun use (â€œNY SAFE Actâ€ n.d.). Therefore, it is possible, though unable to be conï¬rmed based on data in the present study, that such attitudes were a by-product of the legislative changes in the state. On the opposite dimension of concernâ€”prevention, which incorporates gun control-related measuresâ€”these same regions again signiï¬cantly differed. Compared to the Midwest and West, respondents in the Southwest region were signiï¬cantly less supportive of imposing measures that would limit their gun 446 H. JAYMI ELSASS ET AL. rights. Again, this is likely due, at least in part, to the culture of ï¬rearms and prevalence of ownership. In 2019, for example, Texas had the highest number of registered weapons of any state in the nationâ€”and nearly nine times that of New York (Statista 2020), with approximately 36 percent of residents in 2017 registered as gun owners (CBS News 2017). New York, on the other hand, has one of the lowest rates of ownership in the nationâ€”just 10 percent of residents are registered gun owners (CBS News 2017). Despite such disparities in ownership and registered weapons, however, respondents in the Northeast also were less supportive of preventative measures by way of gun control compared to the Midwest and West. Whereas the Southwestern respondents may have objected to having their gun rights restricted, the Northeastern respondents potentially viewed additional restrictions as unnecessary because the measures (background checks and enhanced penalties) already were in place as a result of the SAFE Act. For both disproportionality and hostility, respondents in the Northeast, Southeast, and Midwest were signiï¬cantly more likely to express agreement to related questions as compared to individuals in both the Southwest and West. Related to disproportionality, or the belief that school shootings are happening more than they really are, it is possible that geographic proximityâ€”such that the Northeastern, Southeastern, and Midwestern campuses are spatially closer to Newtown, CT, where the Sandy Hook shooting occurredâ€”played a role in cultivating such differences (see also Dor e et al. 2015), though the Southwestern survey was completed prior to the attack. At the same time, it also is possible that such differences are the results of other factors, such as media consumption habits (e.g., Althaus et al. 2009). Further testing, however, would be needed to conï¬rm such a hypothesis. Still, the fact that these three groups of respondents perceived these events as happening more frequently also may explain their signiï¬cantly greater levels of support on the hostility attribute. In other words, the more respondents believed school shootings were frequently occurring, the more likely they were to express punitive attitudes aimed at curbing such behaviors or punishing offenders. Moreover, for the Midwest sample speciï¬cally, these greater beliefs associated with the disproportionality and hostility attributes also can serve to explain why respondents were more likely to subscribe to the overall notion of moral panic over mass shootings. It is unclear, however, why the Northeast and Southeast did not exhibit similar subscriptions. Such inquiry should be the focus of future research. Finally, related to volatility, or the idea that attention to school shootings is particularly short-lived, respondents from the West were more likely to express agreement than those in the Midwest. Notably, mass shootings more broadly are more likely to occur in the Western regions of the United States (Schildkraut and MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME 447 Elsass 2016), so such attitudes may stem from having a more direct, rather than mediatized, experience with these events. Mixed results were found when considering other potential correlates of moral panic subscription beyond just location, including demographics, fear of crime, and perceived risk of victimization. Compared to the defense dimension of the concern attribute, the three predictors in Schildkraut, Elsass, and Staffordâ€™s (2015) study were again signiï¬cant in the current assessment: females, blacks, and non-gun owners were less likely to express attitudes favorable to gun rights. Similarly, females also were less likely to express agreement that school shootings were unpredictable (volatility), and fear of personal crime was positively correlated with greater attitudes indicative of disproportionality and hostility. Despite these consistencies, however, additional demographic measures were found to be signiï¬cant predictors across all attributes and the general moral panic scale that were not detected in the original study. Given that one of the signiï¬cant changes in the OLS models was the introduction of location as controls, this suggests that, in the context of moral panic over school shootings, there may be regional social processes at work potentially inï¬‚uencing such attitudes. Future research, however, would be needed to conï¬rm such a hypothesis. Taken together, the ï¬ndings of the present study broaden our collective understanding about how people think about school shootings in general as well as through the lens of moral panic theory. This line of inquiry has not only conï¬rmed the employment of a quantitative application of the attributional model to understand the potential for a school shooting moral panic, but also creates additional avenues for future research. For example, while it is possible that the variation in attitudes about school shootings and moral panic are shaped by different, location-speciï¬c social and cultural norms and processes, these cannot fully be captured within the scope of our study. In other words, while differences were found between the regions, we cannot for certain say why those disparities are present. Accordingly, future research may wish to further investigate these processes using different methodologies. Additional consideration may be given to other predictors, such as political ideology or partisanship, prior victimization or offending, and media consumption (amount and/or sources; see generally Althaus et al. 2009; Elsass et al. 2014) that also may predict attitudes associated with moral panic over school shootings. Moreover, given that both quantitative applications of the theory have utilized samples of college students, an expansion of the research to the general public is particularly warranted. While the present study yields a number of important ï¬ndings about individual-level predictors of attitudes related to moral panic over mass shootings, there also are several additional limitations that must be considered. First, this 448 H. JAYMI ELSASS ET AL. study relies on convenience samples at each of the ï¬ve institutions, which limit the generalizability of the results. Similarly, given that only a single university in each region was surveyed, it is not necessarily the case that the ï¬ndings would be generalizable to all college students in that particular region. Replicating the study with an even broader sample would help to improve the generalizability of the ï¬ndings, while continuing to validate the use of a quantitative application of Goode and Ben-Yehudaâ€™s (1994a, 1994b) attributional model of moral panic theory as it relates to school shootings. Despite their statistical rarity, school shootings elicit extreme reactions from members of the general public, who often fear for their own safety despite their unlikelihood of victimization. As a result, the demand for responses often is rooted in emotion rather than evidence. Examining peopleâ€™s personal reactions to social problems has been helpful in the past, although such research typically has relied on historical, qualitative data. Assessing regional differences in individualsâ€™ reactions in real time to a volatile phenomenon such as school shootings is a vital next step that will undoubtedly pay dividends in contextualizing the demand for future legislation. ENDNOTES *Please direct correspondence to Jaclyn Schildkraut, State University of New York at Oswego, 458 Mahar Hall, Oswego, NY 13126, USA; e-mail: Jaclyn.Schildkraut@oswego.edu 1 Using factor analysis to determine which attribute was the best ï¬t for each of the 19 questions used to assess moral panic, the researchers identiï¬ed two opposite dimensions of concern: (1) prevention, incorporating questions representing attitudes towards gun control and (2) defense, including opinions favorable towards gun rights. 2 The number of classes varied by institution based on class enrollments. The minimum number of classes sampled was ï¬ve (Southwest), and the maximum number was ten (West). 3 Due to IRB restrictions, the researchers were not permitted to ask respondents the numeric value of their age, but instead were required to ask respondents where they fell in the following age groups: 18-21, 22-25, 26-29, 30 and older. These were then collapsed into 21 and under and 22 and over categories. 4 Grange and Ferraro (1989) included an eleventh itemâ€”rowdy youth. This was later omitted by Ferraro (1995), the model which was used in the present study. REFERENCES Allison, Paul. 2014. â€œListwise Deletion: Itâ€™s NOT Evil.â€ Retrieved Oct. 5, 2020. . Althaus, Scott L., Anne M. Cizmar, and James G. Gimpel. 2009. â€œMedia Supply, Audience Demand, and the Geography of News Consumption in the United States.â€ Political Communication 26(3):249â€“77. MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME 449 Altheide, David L. 2009a. â€œMoral Panic: From Sociological Concept to Public Discourse.â€ Crime Media Culture 5(1):79â€“99. Altheide, David L. 2009b. â€œThe Columbine Shootings and the Discourse of Fear.â€ American Behavioral Scientist 52(10):1354â€“70. Burns, Ronald and Charles Crawford. 1999. â€œSchool Shootings, the Media, and Public Fear: Ingredients for a Moral Panic.â€ Crime, Law and Social Change 32(2):147â€“68. CBS News. 2017. â€œGun ownership by stateâ€. September 13. Retrieved June 6, 2020. . Chyi, Hsiang I. and Maxwell McCombs. 2004. â€œMedia Salience and the Process of Framing: Coverage of the Columbine School Shooting.â€ Journalism and Mass Communication Quarterly 81(1):22â€“35. Cohen, Stanley. 1972. Folk Devils and Moral Panics: The Creations of the Mods and the Rockers. London: MacGibbon and Kee. Cornell, Dewey. 2015. â€œOur Schools are Safe: Challenging the Misperception that Schools are Dangerous Places.â€ American Journal of Orthopsychiatry 85(3):217â€“20. Crawford, Charles and Ronald Burns. 2015. â€œPreventing School Violence: Assessing armed Guardians, School Policy, and Context.â€ Policing: An International Journal of Police Strategies and Management 38(4):631â€“47. Critcher, Chas. 2008. â€œMoral Panic Analysis: Past, Present, and Future.â€ Sociology Compass 2 (4):1127â€“44. Dizard, Jan E., Robert Muth, and Stephen P. Andrews (eds.). 1999. Guns in America: A reader New York: New York University Press. Dor e, Bruce, Leonard Ort, Oï¬r Braverman, and Kevin N. Ochsner. 2015. â€œSadness Shifts to Anxiety Over Time and Distance from the National Tragedy in Newtown, Connecticut.â€ Psychological Science 26(4):363â€“73. Downs, Anthony. 1972. â€œUp and Down with Ecology: the â€˜Issue-Attention Cycleâ€™.â€ Public Interest 28:38â€“50. Elsass, H. Jaymi, Jaclyn Schildkraut, and Mark C. Stafford. 2014. â€œBreaking News of Social Problems: Examining Media Consumption and Student Beliefs About School Shootings.â€ Criminology, Criminal Justice, Law & Society 15(2):31â€“42. Ferraro, Kenneth F. 1995. Fear of Crime: Interpreting Victimization Risk Albany, NY: State University of New York Press. Ferraro, Kenneth F. and Randy L. Grange. 1987. â€œThe Measurement of Fear of Crime.â€ Sociological Inquiry 57(1):70â€“101. Fox, James A. and Monica J. DeLateur. 2014. â€œMass Shootings in America: Moving Beyond Newtown.â€ Homicide Studies 18(1):125â€“145. Fox, James A. and Emma E. Friedel. 2018. â€œThe Menace of School Shootings in America: Panic and Overresponseâ€. Pp. 15â€“36 in The Wiley Handbook on Violence in Education: Forms, Factors, and Preventions, edited by Harvey Shapiro. New York: Wiley/Blackwell Publishers. Friedel, E. E. 2020. â€œComparing the Impact of Household Gun Ownership and Concealed Carry Legislation on the Frequency of Mass Shootings and Firearms Homicide.â€ Justice Quarterly. https://doi.org/10.1080/07418825.2020.1789693 Frymer, Benjamin. 2009. â€œThe Media Spectacle of Columbine: Alienated Youth as an Object of Fear.â€ American Behavioral Scientist 52(10):1387â€“404. Goode, Erich and Nachman Ben-Yehuda. 1994a. â€œMoral Panics: Culture, Politics, and Social Construction.â€ Annual Review of Sociology 20:149â€“71. Goode, Erich and Nachman Ben-Yehuda. 1994b. Moral Panics: The Social Construction of Deviance Malden, MA: Blackwell Publishing. Graber, Doris. 1980. Crime News and the Public Chicago, IL: University of Chicago Press. 450 H. JAYMI ELSASS ET AL. Graf, Nikki. 2018. â€œA Majority of U.S. Teens Fear a Shooting Could Happen at Their School, and Most Parents Share Their Concernâ€. Pew Research Center, April 18. Retrieved June 6, 2020. . Grange, Randy L. and Kenneth F. Ferraro. 1989. â€œAssessing Age and Gender Differences in Perceived Risk and Fear of Crime.â€ Criminology 27(4):697â€“719. Hall, Stuart, Chas Critcher, Tony Jefferson, John Clarke, and Brian Roberts. 1978. Policing the Crisis: Mugging, the State, and Law and Order. Houndmills: Palgrave Macmillan. Hay, Carter. 1995. â€œMobilization Through Interpellation: James Bulger, Juvenile Crime, and the Construction of a Moral Panic.â€ Social Legal Studies 4(2):197â€“223. Hesterman, Jennifer. 2015. Soft Target Hardening: Protecting People from Attack. Boca Raton, FL: CRC Press. Hier, Sean P. 2002. â€œRaves, Risk, and the Ecstasy Panic: A Case Study in the Subversive Nature of Moral Regulation.â€ The Canadian Journal of Sociology 27(1):33â€“57. Holland, Kristin M., Jeffrey E. Hall, Jing Wang, Elizabeth M. Gaylor, Linda L. Johnson, Daniel Shelby, Thomas R. Simon, and School-Associated Violent Deaths Study Group. 2019. â€œCharacteristics of School-Associated Youth Homicides United States, 1994â€“2018.â€ Morbidity and Mortality Weekly Report 68(3):53â€“60. Jones, Brian J., Joseph A. McFalls, and Bernard J. Gallagher. 1989. â€œToward a Uniï¬ed Model for Social Problems Theory.â€ Journal for the Theory of Social Behavior 19(3):337â€“56. Kalish, Rachel and Michael Kimmel. 2010. â€œSuicide by Mass Murder: Masculinity, Aggrieved Entitlement, and Rampage School Shootings.â€ Health Sociology Review 19(4):451â€“64. Klocke, Brian V. and Glenn W. Muschert. 2010. â€œA Hybrid Model of Moral Panics: Synthesizing the Theory and Practice of Moral Panic Research.â€ Social Compass 4(5):295â€“309. Krinsky, Charles. 2013. â€œIntroduction: The Moral Panic Concept.â€ Pp. 1â€“14 in The Ashgate Research Companion to Moral Panics, edited by Charles Krinsky. Burlington, VT: Ashgate Publishing. Kupchik, Aaron. 2010. Homeroom Security: School Discipline in an Age of Fear. New York: New York University Press. Larkin, Ralph W. 2007. Comprehending Columbine Philadelphia: Temple University Press. Larkin, Ralph W. 2009. â€œThe Columbine Legacy: Rampage Shootings as Political Acts.â€ American Behavioral Scientist 52(9):1309â€“26. Lemieux, Frederic. 2014. â€œEffect of Gun Culture and Firearm Laws on Gun Violence and Mass Shootings in the United States: A Multi-Level Quantitative Analysis.â€ International Journal of Criminal Justice Sciences 9(1):74â€“93. Madï¬s, Eric. 2016. â€œâ€˜Itâ€™s Better to Overreactâ€™: School Ofï¬cialsâ€™ Fear and Perceived Risk of Rampage Attacks and the Criminalization of American Public Schools.â€ Critical Criminology 24(1):39â€“55. Madï¬s, Eric. 2020. How to Stop School Rampage Killing: Lessons from Averted Mass Shootings and Bombings. New York, NY: Palgrave Macmillan. Maguire, Brendan, Georgie Ann Weatherby, and Richard A. Mathers. 2002. â€œNetwork News Coverage of School Shootings.â€ The Social Science Journal 39(3):465â€“70. McCarthy, Justin. 2015. Three in 10 U.S. Parents Worry About Childâ€™s Safety at School. Gallup, August 25. Retrieved June 6, 2020. . McCombs, Maxwell and Jian-Hua Zhu. 1995. â€œCapacity, Diversity, and Volatility of the Public Agenda: Trends from 1954 to 1994.â€ Public Opinion Quarterly 59(4):495â€“525. McCorkle, Richard C. and Terance D. Miethe. 1998. â€œThe Political and Organizational Response to Gangs: An Examination of a â€˜Moral Panicâ€™ in Nevada.â€ Justice Quarterly 15(1):41â€“64. MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME 451 Miller, Toby. 2013. â€œTracking moral panic as a concept.â€ Pp. 37â€“54 in The Ashgate Research Companion to Moral Panics, edited by Charles Krinsky. Burlington, VT: Ashgate Publishing. Muschert, Glenn W. 2002. â€œMedia and Massacre: The Social Construction of the Columbine Story.â€ PhD dissertation, Department of Sociology, University of Colorado at Boulder. Muschert, Glenn W. 2007a. â€œResearch in School Shootings.â€ Sociology Compass 1(1):60â€“80. Muschert, Glenn W. 2007b. â€œThe Columbine Victims and the Myth of the Juvenile Superpredator.â€ Youth Violence and Juvenile Justice 5(4):351â€“66. Muschert, Glenn W. 2009. â€œFrame-Changing in the Media Coverage of a School Shooting: The Rise of Columbine as a National Concern.â€ The Social Science Journal 46(1):164â€“70. Muschert, Glenn W. and Ralph W. Larkin. 2007. â€œThe Columbine High School shootings.â€ Pp. 253â€“266 in Crimes and Trials of the Century, edited by Steven Chermak and Frankie Y. Bailey. Westport, CT: Praeger. Muschert, Glenn W. and Massimo Ragnedda. 2010. â€œMedia and Violence Control: The Framing of School Shootingsâ€. Pp. 345â€“361 in The Control of Violence in Modern Society: Multidisciplinary Perspectives, From School Shootings to Ethnic Violence, edited by Wilhelm Heitmeyer, Heinz-Gerhard Haupt, Stefan Malthaner, and Andrea Kirschner. New York: Springer Publishing. National Conference of State Legislatures. 2019. â€œGuns on Campus: Overview.â€ Retrieved June 6, 2020. . Newman, Katherine. 2006. â€œSchool Shootings are a Serious Problem.â€ Pp. 10â€“17 in School Shootings, edited by Susan Hunnicutt. Farmington Hills, MI: Greenhaven Press. Newman, Benjamin J. and Todd K. Hartman. 2019. â€œMass Shootings and Public Support for Gun Control.â€ British Journal of Political Science 49(4):1527â€“53. â€œNY SAFE Act.â€ n.d. Albany, NY: State of New York. Retrieved June 6, 2020. . Pew Research Center for the People and the Press. 1999. â€œColumbine Shooting Biggest News Draw of 1999â€. December 28. Retrieved June 6, 2020. . Pew Research Center for the People and the Press. 2007. â€œWidespread Interest in Virginia Tech Shootings, but Public Paid Closer Attention to Columbineâ€. April 25. Retrieved June 6, 2020. . Robinson, Matthew B. 2011. Media Coverage of Crime and Criminal Justice. Durham, NC: Carolina Academic Press. Rothe, Dawn and Stephen L. Muzzatti. 2004. â€œEnemies Everywhere: Terrorism, Moral Panic, and US Civil Society.â€ Critical Criminology 12(3):327â€“50. Saad, Lydia. 1999. â€œPublic Views Littleton Tragedy as Sign of Deeper Problems in Countryâ€. Gallup News, April 23. Retrieved June 6, 2020. . Schildkraut, Jaclyn. 2014. â€œMass Murder and the Mass Media: An Examination of the Media Discourse on U.S. Rampage Shootings, 2000â€“2012â€. PhD dissertation, School of Criminal Justice, Texas State University. Schildkraut, Jaclyn. 2016. â€œMass Murder and the Mass Media: Understanding the Construction of the Social Problem of Mass Shootings in the U.S.â€ Journal of Qualitative Criminal Justice and Criminology 4(1):1â€“41. Schildkraut, Jaclyn and Collin M. Carr. 2020. â€œMass Shootings, Legislative Responses, and Public Policy: An Endless Cycle of Inactionâ€. Emory Law Journal 69(5), 1043â€“76. Schildkraut, Jaclyn, Margaret K. Formica, and Jim Malatras. 2018. Can Mass Shootings be Stopped? To Address the Problem, we Must Better Understand the Phenomenon (Policy 452 H. JAYMI ELSASS ET AL. brief). Retrieved June 6, 2020. . Schildkraut, Jaclyn and Tiffany C. Hernandez. 2014. â€œLaws That Bit the Bullet: A Review of Legislative Responses to School Shootings.â€ American Journal of Criminal Justice 39(2):358â€“ 74. Schildkraut, Jaclyn and H. Jaymi Elsass. 2016. Mass Shootings: Media, Myths, and Realities. Santa Barbara, CA: Praeger. Schildkraut, Jaclyn, H. Jaymi Elsass, and Kimberly Meredith. 2018. â€œMass Shootings and the Media: Why All Events are Not Created Equal.â€ Journal of Crime and Justice 41(3):223â€“243. Schildkraut, Jaclyn, H. Jaymi Elsass, and Glenn W. Muschert. 2016. â€œSatirizing Mass Murder: What Many Think, Yet Few Will Sayâ€. Pp. 233â€“255 in Gun Violence in American Society: Crime, Justice, and Public Policy, edited by Lisa Eargle, and Ashraf Esmail. Lanham, MD: University Press of America. Schildkraut, Jaclyn, H. Jaymi Elsass, and Mark C. Stafford. 2015. â€œCould it Happen Here? Moral Panic, School Shootings, and Fear of Crime Among College Students.â€ Crime, Law and Social Change 63(1â€“2):91â€“110. Schildkraut, Jaclyn and Glenn W. Muschert. 2013. â€œViolent Media, Guns, and Mental Illness: The Three Ring Circus of Causal Factors for School Massacres, as Related in Media Discourse.â€ Fast Capitalism 10(1). Retrieved from . Schildkraut, Jaclyn and Glenn W. Muschert. 2014. â€œMedia Salience and the Framing of Mass Murder in Schools: A Comparison of the Columbine and Sandy Hook massacres.â€ Homicide Studies 18(1):23â€“43. Schildkraut, Jaclyn and Glenn W. Muschert. 2019. Columbine, 20 Years Later and Beyond: Lessons from Tragedy. Santa Barbara, CA: Praeger. Soraghan, Mike. 2000. â€œColorado After Columbine: The Gun Debate.â€ State Legislatures 26(6):14â€“ 21. Springhall, John. 1999. â€œViolent Media, Guns, and Moral Panics: The Columbine High School Massacre, 20 April 1999.â€ Paedagogica Historica 35(3):621â€“41. Statista. 2020. â€œNumber of Registered Weapons in the U.S. in 2019, by Stateâ€. Retrieved June 6, 2020. . Surette, Ray. 2015. Media, Crime, and Criminal Justice: Images and Reality, 5th ed. Paciï¬c Grove, CA: Brooks/Cole Publishing Company. Tavakol, Moshen and Reg Dennick. 2011. â€œMaking Sense of Cronbachâ€™s Alpha.â€ International Journal of Medical Education 2(1):53â€“55. Tewksbury, Richard and Elizabeth E. Mustaine. 2003. â€œCollege Studentsâ€™ Lifestyles and SelfProtective Behaviors: Further Considerations of the Guardianship Concept in Routine Activities Theory.â€ Criminal Justice and Behavior 30(3):302â€“27. Tonso, Karen L. 2009. â€œViolent Masculinities as Tropes for School Shooters: The Montreal Massacre, the Columbine Attack, and Rethinking Schools.â€ American Behavioral Scientist 52 (9):1266â€“85. Vasilogambros, Matt. 2018. â€œAfter Parkland, States Pass 50 New Gun-Control Laws.â€ Pew Research Center, August 2. Retrieved June 6, 2020. . Waddington, P. A. J. 1986. â€œMugging as a Moral Panic: A Question of Proportion.â€ British Journal of Sociology 37(2):245â€“59. Warr, Mark. 1984. â€œFear of Victimization: Why are Women and the Elderly More Afraid?â€ Social Science Quarterly 65(3):681â€“702. MORAL PANIC, SCHOOL SHOOTINGS, AND FEAR OF CRIME 453 Warr, Mark. 2000. â€œFear of Crime in the United States: Avenues for Research and Policy.â€ in Measurement and Analysis of Crime: Criminal Justice 2000, edited by David Duffee. Washington, DC: US Department of Justice, Ofï¬ce of Justice Programs. Welch, Michael, Eric A. Price, and Nana Yankey. 2002. â€œMoral Panic Over Youth Violence: Wilding and the Manufacturing of Menace in the Media.â€ Youth and Society 34(1):3â€“30. Wike, Traci L. and Mark W. Fraser. 2009. â€œSchool Shootings: Making Sense of the Senseless.â€ Aggression and Violent Behavior 14(3):162â€“69. Young, Jock. 1971. â€œThe Role of the Police as Ampliï¬ers of Deviancy, Negotiators of Reality and Translators of Fantasy: Some Consequences of our Present System of Drug Control as Seen in Notting Hill.â€ Pp. 27â€“61 in Images of Deviance, edited by Stanley Cohen. Harmondsworth, UK: Penguin Books. Zatz, Marjorie S. 1987. â€œChicano Youth Gangs and Crime: The Creation of a Moral Panic.â€ Crime, Law and Social Change 11(2):129â€“58. Appendix Moral Panic Indicator Likert Questions with Descriptive Statistics 1: Variable Questions Min. Max. Mean Std. Dev. Consensus 3 15 10.87 1.885 School shootings are premeditated events I feel safe on campus Attention to a school shooting event does not last very long before another news story replaces it Concern (Defense) 2 10 5.05 2.692 I believe professors should be able to carry a ï¬rearm on campus if they have a concealed handgun license I believe students should be able to carry a ï¬rearm on campus if they have a concealed handgun license Concern (Prevention) 2 10 8.14 1.752 I believe that people should have to pass a criminal background check to purchase a ï¬rearm from a private dealer or a gun show I believe that there should be very stiff penalties assessed for guardians who fail to keep ï¬rearms out of the reach of their minor children Hostility 4 20 14.45 3.699 I believe my school should have more rigorous safety procedures in place I believe that someone who brings a ï¬rearm to school should be expelled I believe that someone who brings a ï¬rearm to school should serve time in jail 454 H. JAYMI ELSASS ET AL. 1 (continued) Variable Questions Min. Max. Mean Std. Dev. I believe that someone who brings a ï¬rearm to school should receive a mental health evaluation Disproportionality 6 30 19.32 4.753 School shootings are a major problem in the United States I believe school shootings are a major issue worldwide I believe a school shooting could happen at my school I believe that juvenile crime over the last 20 years is on the rise I believe that school shootings are on the rise The odds of being a victim of a school shooting are greater than being struck by lightning Volatility 2 10 6.52 1.800 School shootings are unpredictable I believe the media devotes too much coverage to school shootings MP General Scale 19 95 64.43 8.707 View publication stats Purchase answer to see full attachment