Abstract
We examine if adverse childhood experiences (ACEs) directly affect the amount of harm (victims and injuries) caused in mass public shootings or whether ACEs have a mediating or moderating influence on a variety of factors previously associated with the amount of harm caused in mass shootings. Using publicly available data, our results show that ACEs do not directly affect harm. In the mediation model, results indicate the number of ACEs experienced by the shooters indirectly affected the harm of the event, but only through the shooters possessing fame-seeking motivations. In the moderation model, the significant interactions show the strongest relationships between ACEs and harm for those without mental distress or life stressors or signs of crisis. Recommendations for future research and practice are offered.
Mass public shootings have increased and become more deadly in the 21st century (Duwe, 2020; Krouse & Richardson, 2015; Lankford & Silver, 2020; Schildkraut & Turanovic, 2022). From 1966 to December 2023, there have been 193 mass public shootings resulting in 1,391 deaths (The Violence Project, 2023). The impact of these events is widespread, affecting not only the family and friends of the victims but also affecting the mental health of survivors and members of the involved communities (Lowe & Galea, 2017).
Recent research found that many of these mass public shooters have experienced traumatic events in their childhood (M. Sanchez & Ferguson, 2022). Social workers have long known that adverse childhood experiences (ACEs) can lead to a variety of negative adult behavioral outcomes including involvement in general criminal activity and violent behaviors (Baglivio et al., 2015; Bellis et al., 2014; Teague et al., 2008), reduced mental health (Merrick et al., 2017), and lower distress tolerance skills (Rudenstine et al., 2019). Furthermore, the effect of ACEs appears to be cumulative, with individuals experiencing more ACEs having greater detrimental outcomes (Bellis et al., 2014; Duke et al., 2010; Kim et al., 2021).
The purpose of this research is to examine the role of ACEs in the perpetrator’s life to predict the number of victims (killed and injured), that is, harm, caused by mass public shooting events. Using data from the Violence Project (Peterson & Densley, 2022), a publicly available dataset, we examine if ACEs identified in the perpetrator’s life directly affect the amount of harm caused by the mass public shooting or whether ACEs have a mediating or moderating influence on a constellation of risk factors previously associated with the harm in mass shootings. A mass public shooting is defined in this dataset as a:
. . .a multiple homicide incident in which four or more victims are murdered with firearms—not including the offender(s)—within one event, and at least some of the murders occurred in a public location or locations in close geographical proximity (e.g., a workplace, school, restaurant, or other public settings), and the murders are not attributable to any other underlying criminal activity or commonplace circumstance (armed robbery, criminal competition, insurance fraud, argument, or romantic triangle). (Krouse & Richardson, 2015, p. 10)
Understanding the impact of ACEs on those who perpetrate mass public shootings has the potential to assist social workers in identifying potential offenders and intervening to support these individuals and possibly avert tragedy (Segeren et al., 2020). Furthermore, this research will contribute to the sparse, albeit growing, social work research focused on gun violence, particularly on the topic of mass public shootings, a reoccurring tragedy played out across the United States (Sperlich et al., 2022).
Literature Review
Mass public shootings, although they represent only a small percentage of all homicides, invoke fear and dread among citizens in the United States (C. Sanchez et al., 2020). Although the phenomenon is not unique to the United States, mass murders in the United States have a number of distinct features, including that they are much more likely to involve firearms and result in higher fatality rates (National Council for Behavioral Health, 2019). A review of the literature on ACE occurrences and risk factors for gun violence, specifically mass public shootings, helped to inform the conceptualization of this study. We present an overview of this scholarship next.
ACEs and Mass Shooters
Recently, researchers in a variety of fields have recognized the deleterious impacts of ACEs on a variety of behaviors, including violent crimes. Briefly, the ACEs study conducted by Felitti et al. (1998) explored health outcomes related to the categories of childhood distress, family dysfunction, and types of abuse. They discovered a graded relationship between the number of ACEs and adult health risks and behaviors, including alcohol abuse, drug abuse, depression, suicide attempts, smoking, ischemic heart disease, cancer, chronic lung disease, skeletal fractures, and liver disease. Most adults experience at least one type of ACE, with approximately 16% experiencing more than four ACEs (Merrick et al., 2017).
Researchers examining the presence of ACEs among mass shooters found that a majority of shooters had at least one type of ACE and that many had several ACEs in their backgrounds. Examining mass shooters who acted alone, Gill et al. (2017) found that 63% of their sample had experienced long-term stress which included being a victim of abuse during childhood. Peterson and Densley (2019) also found that almost one third of their sample of 167 mass public shooters had experienced severe childhood trauma. Alathari et al. (2019) studying school shooters found an average of 3.4 ACEs for the shooters in their sample and that almost every shooter had experienced some type of ACE. Similarly, Dowdell et al. (2022) reported that 72% of the 25 school shooters they examined had experienced at least one ACE. Finally, Silver and Silva (2022) examined the deadliest mass public shootings in an 11-year period from 1999 to 2020 and found that childhood and family problems were indirectly related to these deadly attacks and that family problems were a particularly salient factor.
As ACE screenings and knowledge of these factors have been integrated into general health and social–behavioral care settings, there is increasing attention to the limitations of this framework. There are recent efforts to expand the descriptions of ACE exposures to include consideration of additional individual, household, and community-level adverse experiences, including the impact of intergenerational racial stress and trauma (Dube, 2018; Finkelhor et al., 2015; Wade et al., 2016). Dube (2018) argues the omission of these experiences is common among health care professionals who carry out ACE screenings, often using a “biomedical or the etic view of health” while minimizing the impact of the “subjective” life experiences of the patient (p. 181). Yet, researchers note that expanding the categories of ACEs conflicts with empirical methods for supporting valid measurements (Portwood et al., 2021).
For example, Portwood et al. (2021) note that using a broad definition of ACE categories during an assessment is most useful when screening individuals for additional evaluation, while narrow definitions of ACE categories during a screening are preferable when referring individuals for further services. These authors also acknowledge that individuals being assessed for ACEs may view these “risk factors” as positive experiences in their lives (p. 184).
In addition, there are those who argue for caution regarding widespread ACE screening without appropriate interventions in place to respond to disclosures (McLennan et al., 2024). Given the increased attention to the limitations of the original ACE framework, researchers are continuing to examine what should be included in an ACE assessment, how this should be assessed, and what are effective services to remedy the result of the screening (Portwood et al., 2021). Notwithstanding the critiques of ACEs, it remains a viable framework for understanding childhood adversities in the lives of mass shooters (Dowdell et al., 2022; Gill et al., 2017; M. Sanchez & Ferguson, 2022).
Risk Factors Associated With Mass Public Shooters
Several explanations have been advanced to explain the widespread frequency of gun violence, including mass public shootings. These factors include the background characteristics of the shooters (Silver & Silva, 2022; Sutton, 2020), the mental health of the offenders (Duwe, 2020; Lankford & Cowan, 2020), the availability of firearms (Anisin, 2018; Green-Colozzi & Silva, 2022; Sutton, 2020), and the widespread media coverage of such events generate suggesting celebrity status for these shooters (Lankford & Madfis, 2018). The factors frequently associated with this extreme form of violent behavior are also outcomes that are commonly associated with experiencing ACEs during childhood. Below we briefly discuss those issues most noted as risk factors in the mass shooting literature, particularly those related to the number of victims shot, that is, the amount of harm caused during the event. In addition, we discuss the research relating ACEs to each of these factors.
Mental Illness
The association between ACEs and having a mental illness diagnosis is identified throughout ACE literature (Davies et al., 2022; Giano et al., 2021; Loeb et al., 2022; Stinson et al., 2016). In a literature review of two decades of ACE research, Zarse et al. (2019) indicated a connection between ACEs and a wide range of psychiatric illnesses, including depression, posttraumatic stress disorder, and psychosis. A strong correlation additionally exists between ACEs and high levels of anxiety as well as a reduced quality of life (Davies et al., 2022). Finally, higher rates of overall ACEs are reported by individuals who struggle with panic disorders (Zhang et al., 2023).
The relationship between mental health issues and mass shootings is more complicated, in part because media accounts of mass shootings frequently overemphasize the role of mental illness in the etiology of the event (Aitken et al., 2008; McGinty et al., 2016). Some researchers have noted that a majority of mass public shooters have either shown signs of mental illness or had been diagnosed with a mental illness prior to the attack (Duwe, 2020; Lankford & Cowan, 2020). In contrast, other researchers believe that there is only a modest connection between serious mental illness and mass shootings (Fox & Fridel, 2016; National Council for Behavioral Health, 2019; Skeem & Mulvey, 2020). Lin et al. (2018) examined this relationship at the aggregate level and concluded that state-level rates of serious mental illness were not related to mass shooting rates. Although the relationship between ACEs and mental illness is well documented, the relationship between mental illness and mass shootings is less clear.
Suicidal Ideation
Suicidal ideation was identified as one of the key issues associated with the experience of ACEs in the initial ACE study (Felitti et al., 1998). Fuller-Thomson et al. (2016) surveyed 22,559 adults regarding their experience with ACEs, suicide attempts, and other mental health factors. The authors found that the prevalence of suicide attempts was higher among those who experienced specific ACEs versus those who had not experienced those specified ACEs. They discovered participants who reported childhood physical abuse (CPA) reported attempted suicide prevalence rates of 12.40% versus 1.93% of those who had not experienced CPA; those who reported childhood sexual abuse (CSA) had attempted suicide prevalence rates of 16.89% versus 2.09% of participants who had not experienced CSA; and those who reported witnessing parental domestic violence (PDV) had attempted suicide prevalence rates of 17.34% versus 2.32% of participants who had not experienced PDV. Although experiencing just one ACE increases the risk of suicidal ideation, Thompson and Kingree (2022) note that suicidal attempts increase as the number of ACEs increases.
The association between suicidal ideation and mass shooters has also been well documented (Lankford et al., 2021; Peterson & Densley, 2019; Silver et al., 2018; Sutton, 2020). In fact, Silver et al. (2018) found that almost half of the active shooters they studied had suicidal ideations prior to the shootings. Many of these shooters intended to die during the event, either by planning on being shot by police (suicide-by-cop) or by committing suicide during or after the event (Holmes & Holmes, 1992; Kelly, 2010; Krouse & Richardson, 2015).
Lankford et al. (2021) observe that the act of committing a mass public shooting should be considered a “defacto suicide attempt” (p. 153) because even those shooters who survive face either life in prison or execution. They found that mass public shooters were significantly more likely than both the general public and other homicide offenders to have suicidal ideations. Furthermore, there is also a relationship between the number of victims and the likelihood that the offender would die as a result of the mass shooting (Lankford, 2015). Lankford and Cowan (2020) found that among the most lethal offenders they studied, all of them were either suicidal or demonstrated signs of mental illness.
Personal Crisis and Recent Stressors
Childhood adversities tend to be interconnected and can harm a developing child’s brain, which can promote maladaptive health and behavioral patterns (Fogel et al., 2024; McLaughlin, 2016). This likely contributes to the connection between ACEs and an increase in adult stressors and emotional crises (Solberg et al., 2024). Higher ACE scores have been associated with lower socioeconomic status and increased adult adversity, including incarceration, one or more periods of homelessness, marital relationship loss, involuntary employment loss, and disability status (Nurius et al., 2015).
In examining the factors that might potentially trigger a mass shooter to engage in violence, several researchers have noted that mass shooters tend to experience a personal crisis or stressor before the shooting event (Gill et al., 2017; Lankford, 2018b; Levin & Madfis, 2009; Meloy et al., 2004; Silver et al., 2018). Levin and Madfis (2009) refer to this as acute strain. Among adults, these stressors can include things, such as a job loss, a romantic breakup, or a monetary/financial loss. Among adolescents, this strain might also include bullying (Alathari et al., 2019; Dowdell et al., 2022).
Silver et al. (2018), examining the behaviors and characteristics of active shooters prior to the shooting event, found that they experienced an average of 3.6 stressors in the year before the attack, including mental health stressors, financial strain, job-related stressors, and conflicts with friends and peers. Silver and Silva (2022), focusing on the deadliest shootings from 1999 to 2020, noted that the majority of shooters experienced family problems, lost jobs, physical health issues, or financial problems, and that these occurred close to what they noted were planning and preparation activities for the event. These researchers all found that mass murderers were likely to have experienced some type of stressor in the time period leading up to the attack. Those examining the timing of these events believe that these stressors are precipitating events that might trigger an offender to lash out violently.
Substance Use
A significant amount of literature has associated ACEs with an increased likelihood of substance use in adulthood, although Brown and Shillington (2017) noted that this association may be related to individuals wanting to escape unwanted emotions or their surroundings. Studies have found a graded relationship between the number of ACEs and the risk of alcohol abuse (Loudermilk et al., 2018), drug use, drug addiction, and parenteral drug use (Dube et al., 2003).
Although prominent in the literature examining the impacts of ACEs, very few researchers have assessed the role of substance use in the perpetration of a mass public shooting. Gill et al. (2017) studying solo mass murderers found that 44% of them had a history of substance abuse. Brucato et al. (2022) also found that mass murderers in the United States were more likely to use recreational drugs and/or misuse alcohol (22%). Interestingly, they also found the rates of drug and alcohol use to be higher among mass shooters in the United States versus those in other countries.
Fame-Seeking
Previous research has noted that individuals who experience ACEs may be more likely to seek attention as a means of coping with their childhood experiences and improving their sense of self-worth (Huxley & Bizumic, 2017). ACEs and childhood maltreatment have also been shown to increase the likelihood of narcissism (Gao et al., 2024; Marusak et al., 2015). Although not directly studying fame-seeking motivations, others have noted a connection between narcissism and violent behavior (Aitken et al., 2008; Bushman, 2018; Langman, 2009; Lankford, 2018b; Twenge & Campbell, 2009). Knoll and Annas (2015) observe that mass shooters tend to be narcissistic, depressed, and have issues with self-esteem.
A strong correlation seems to exist between those who seek notoriety or fame from the shooting event and the lethality of the event (Lankford, 2016b, 2018a; Lankford & Silver, 2020), with individuals seeking to achieve infamy from the mass shooting having the desire to kill the largest number of victims possible. Lankford and Silver (2020) note that these shooters often “attempt to compensate for their failures in life by creating legacies that will persist long after their deaths” (p. 32).
The desire to create these legacies is also associated with a desire to kill large numbers of persons (Lankford & Silver, 2020). The lethality of the event is strongly related to the amount of attention these individuals will receive (Duwe, 2004; Schildkraut et al., 2017; Silva & Capellan, 2019; Silva & Greene-Colozzi, 2019; Silva & Lankford, 2024). Mass shooters who are motivated by this desire are responsible for a disproportionate number of causalities and on average kill more than twice the number of victims as other mass shooters (Lankford, 2016b). This has been exacerbated by the extent of coverage these events receive, social media, and a general emphasis on the importance of celebrity in our culture (Knoll & Annas, 2015; Lankford, 2016b, 2018a; Lankford & Madfis, 2017).
Availability of Firearms
In addition to the risk factors discussed above, the number of firearms available is an important factor to consider when discussing the harm caused by mass shootings. Numerous scholars have noted that the availability of firearms is a primary factor in the ability of an offender to carry out their plans, particularly when discussing the lethality of the event (Lankford et al., 2021; Peterson & Densley, 2019; Rocque & Duwe, 2018; C. Sanchez et al., 2020). Mass shooters, who cause the most damage, tend to bring multiple weapons to the scene (Berkowitz et al., 2019; Lankford, 2015, 2016a; Lankford & Silver, 2020; Silva & Greene-Colozzi, 2021). Lankford (2016a) notes that mass public shooters in the United States, in part due to our gun culture, are much more likely than mass public shooters in other countries to use multiple weapons. He goes so far as to suggest that limiting mass shooters to only one weapon would decrease the average number of victims by 15% in these attacks (p. 180).
Current Study
The literature reviewed above draws connections between the ACEs and a variety of negative outcomes that have previously been associated with mass public shootings and their harm. Although there is little evidence that the presence of ACEs in an individual’s life directly affects the death and injury associated with a mass public shooting event, it is likely that the consequences associated with the experience of ACEs might be associated with this harm. Therefore, we hypothesize that ACEs are related to a variety of factors previously identified as risk factors related to the harm of mass shootings. This study seeks to understand the potential link between ACEs and the harm caused by mass public shootings by examining risk factors that might link ACEs to the harm caused by mass public shootings. To do so, we examine whether the risk factors discussed above mediate or moderate the relationship between ACEs and the damage caused by mass shootings.
Both mediation and moderation allow us to examine the nature of a relationship to determine exactly how a set of variables are related. Mediation analysis allows us to test whether or not ACEs indirectly affect the number of victims in a shooting through their influence on a set of mediators. In essence, it tests for the presence of indirect relationships. Moderation posits that the influence of ACEs on the number of victims is dependent on the values of a third variable. For example, examining the relationship between ACEs, mental illness, and the number of victims through mediation theorizes that the number of ACEs influences the presence of a mental illness, which in turn influences the number of victims (ACEs -> mental illness -> victims). If mental illness moderates the relationship between ACEs and victims, that suggests that the nature of the relationship (strength, direction) between ACEs and victims varies depending on whether the shooter had a mental illness. Running both models allows us to tease out the nature of the relationships between ACEs and a variety of risk factors in predicting the number of victims or the harm in mass shootings.
Figure 1 represents the path diagram illustrating the hypothesized relationships between the total number of ACEs experienced by shooters, the various risk factors predicted to be associated with the harm caused by mass public shooting events, and the ultimate harm caused by the event (i.e., the number of victims shot).
Conceptual Path Model Predicting Total Number of Victims in Public Mass Shootings.
Specifically, it is hypothesized that the number of ACEs will influence the chances that a shooter will experience mental health issues, be suicidal, use alcohol/drugs, react inappropriately to stressors, and possess fame-seeking motivations. It is posited that as the total number of ACEs increases, the likelihood that the shooter will experience the issues above also increases. Furthermore, it is hypothesized that increases in these factors will, in turn, increase the harm caused by the shooting event (i.e., the number of victims shot). In other words, mental health issues, suicidal ideation, the use of alcohol/drugs, the number of stressors in the shooter’s life, and fame-seeking motivations will mediate or intervene in the relationship between ACEs and the ultimate harm caused by the event.
Figure 2 illustrates the moderated effects, where it is hypothesized that the total number of ACEs experienced by the shooter will interact with the mental health of the shooter, the shooter’s suicidal ideations, use of alcohol/drugs, potential stressors or triggering events, and fame-seeking motivations. Thus, it is the combination of having experienced a larger number of ACEs and these risk factors that increase the amount of harm caused by the event.
Interaction Model Predicting Total Number of Victims in Public Mass Shootings.
Method
This research uses data from the Violence Project, a comprehensive database compiled by Peterson and Densley (2022) covering mass public shootings in the United States from 1966 to 2022. The final dataset consists of 188 identified shooters for which there is background information on the offender. The data include information on shooters and their backgrounds, and information about the actual shooting. Information was obtained from both primary sources (e.g., journals, suicide notes, interview transcripts) and publicly available secondary sources (e.g., media sources, court transcripts, medical records) (see Peterson & Densley, 2022). The UNC Charlotte Institutional Review Board (IRB) determined that this project did not fall under the category of human subjects and that it did not require IRB approval (IRB 24-0748).
Peterson and Densley (2021) note that for each perpetrator, they often had more than 150 various pieces of information from these various sources for each offender. The coding and interpretation of various information was the subjective decision of the researchers; however, each case was coded by two coders with the results being compared and discrepancies resolved by the principal investigators who also analyzed all of the information to make a decision. In making these decisions, primary sources were given more weight than secondary sources (Peterson & Densley, 2022).
Variables
Exogenous
The presence of ACEs in the offenders’ lives was measured by a cumulative score indicating the total number of ACEs experienced by each offender. ACEs included being raised by a single parent, parental divorce/separation, parent death in childhood, parental suicide, being physically abused, being sexually abused, being emotionally abused, being neglected, witnessing the violent treatment of his or her mother, parental substance use, parental criminal record and/or family member incarcerated. These are the ACEs available in this dataset.
Endogenous
In the mediated path model, it is hypothesized that the mental health of the offender, suicidal ideation, fame-seeking motives, substance use, personal crises, and recent stressors will operate as mediators in the relationship between ACEs and total harm.
Mental health issues are measured by whether the offender had been officially diagnosed with a mental illness (1 = diagnosed). Suicidal ideation was measured by a dichotomous variable, with the one (1) code indicating that the offender had been suicidal prior to the shooting or intended to die during the shooting.
Fame-seeking motivations were also measured by a dichotomous variable, with the one (1) code indicating that the offender possessed fame-seeking motivations. Substance use is also measured as a dichotomous variable with the one (1) code signifying that the offender had a problem with alcohol, marijuana, and/or other drugs.
An index variable was created to measure signs that the shooter was in crisis prior to the shooting. These signs include the inability to perform daily tasks, notably depressed mood, unusually calm or happy, rapid mood swings, increased agitation, abusive behavior, and/or isolation, and each of these is given a score of 1 if the behavior represents a marked change from the offender’s baseline behavior.
Finally, a measure of the number of recent or ongoing stressors or triggering events, including whether the offender had experienced a recent breakup, an employment stressor, an economic stressor, a family issue, a legal issue, been bullied, or experienced some other type of crises, was developed.
Final Endogenous
As we are interested in predicting the ultimate amount of harm caused by mass shooting offenders, total harm is measured by the total number of victims who were shot during the incident (not including the shooter). This includes victims who were killed and those who were injured. To avoid having a highly skewed distribution, the number of injured victims in the case of Stephen Paddock (Las Vegas, October 2017) was restricted from 867 to 70 (the next highest number of victims).
Interactions
Interaction terms were created for the interaction between ACEs and mental health issues, suicidal ideation, fame-seeking motivations, substance use, signs of personal crises, and stressors. Interaction variables were computed by taking the multiplicative effects of the total number of ACEs experienced by each offender times each of the endogenous variables, creating a total of six interaction variables.
Controls
Because of the importance of firearms acquisitions to the perpetration of mass public shootings, particularly those with higher numbers of victims, each of the analysis’s controls for the total number of firearms brought to the scene. Firearms include handguns, shotguns, rifles, and assault weapons. Three shooters brought 10 or more weapons to the scene, skewing the distribution of this variable. The range on firearms brought to the scene was restricted to a maximum number of eight guns.
Sample Characteristics
The sample was overwhelmingly male (97.3%), excluding one non-binary individual who was assigned male at birth. (See Table 1). The majority of the shooters were white (56.9%), 22.3% were Black, and 9% were Latinx. The shooters ranged in age from 11 to 70 years, with an average age of 33.6 years. The total number of ACEs in the sampled shooters ranged between 0 and 9, with the average shooter having experienced one ACE during their childhood. The most common ACEs among this group were parental divorce/separation (23.5%), raised by a single parent (12.8%), emotionally abused (12.3%), physically abused (10.8%), and parent had a criminal record (10.2%). The remaining ACEs were less frequently reported among this group, and include parental substance abuse (8.0%), experienced neglect (7.0%), family member incarcerated (5.3%), sexually abused (5.3%), parent death (4.3%), or suicide (3.2%).
Description of the Sample (n = 188).
The average shooter was experiencing at least some signs of a crisis prior to the shooting (
Notably, 7% of the shooters exhibited fame-seeking motivations and the average shooter brought two firearms to the scene (
Analytical Strategy
All analyses were performed using RStudio version 2021.09.0.351 (RStudio Team, 2021). Missing values were imputed using predictive mean matching (van Burren & Groothuis-Oudshoorn, 2011). Predictive mean matching has been deemed appropriate when the data are not MNAR (missing not at random) and less than 50% of the cases have missing data (Marshall et al., 2010). In our data, the vast majority of the variables were missing less than 2% of their cases, one variable had 9.5% missing and one variable had 15% missing. Little’s MCAR (missing completely at random) test indicated that the missing data were missing completely at random.
To examine the mediation model, the authors ran the path model using Lavaan package (Rosseel, 2012). Both direct and indirect effects were calculated. The Holm–Bonferroni correction was applied to the observed significant relationships post hoc. This did not change the significance of any of the observed relationships.
To examine the moderation effects, the authors ran linear models predicting the total number of victims by both the interaction effect and the main effects, and the number of firearms brought to the scene for each of the interactions. Plots of the interaction effects were generated using the “sjPlot” package in R (Lüdecke, 2021).
Results
Mediation Model
The mediation model tests the hypothesis that experiencing ACEs during childhood increases the chances that a shooter will experience or exhibit a number of influences thought to be associated with the harm caused by mass public shootings. Specifically, we examine whether signs of a mental health crisis, a mental health diagnosis, suicidality, substance use, fame-seeking motivations, and stressors mediate the relationship between one’s cumulative ACE score and the harm of the mass shooting event.
Figure 3 shows the results from the path analysis. The results show that the cumulative number of ACEs the shooter experienced does not have a direct impact on the amount of harm caused by the shooting event. Although the total number of ACEs was shown to weakly increase the likelihood of the shooter having a mental health diagnosis, the diagnosis in turn was not related to the harm of the event. Contrary to our predictions, most of the risk factors examined were not associated with either ACEs or the number of victims. The number of ACEs experienced by the shooters indirectly affected the harm of the event only through the shooters possessing fame-seeking motivations. Shooters who experienced greater numbers of ACEs were more likely to be seeking fame by perpetrating the shooting (β = .23). In addition, shooters with fame-seeking motivations killed and injured greater numbers of victims (β = .33).
Path Analysis Predicting the Total Number of Victims in Public Mass Shootings.
The indirect impact of the ACE score on the harm of the event was .08 (.23 × .33). This means that those with a high number of ACEs were more likely to possess fame-seeking motivations, which in turn increased the number of victims harmed during the event. However, this relationship was very weak. Calculating the indirect effects from the unstandardized coefficients (not shown), we find that every additional ACE that the shooter experiences leads to a .688 increase in victims. So, each increase in two additional ACEs increases the total number of victims by roughly one. Finally, the number of firearms brought to the scene was a weak predictor of harm (β = .29). The evidence does not support the hypothesis that ACEs have a direct or indirect effect on harm, except for its weak influence through fame-seeking motivations.
Overall, the analysis shows that the model is a modest fit for the data. The model explained only 26.0% (R2 = .26) of the variance in the total number of victims, and the majority of that was explained by the number of firearms brought to the scene by the shooter.
Moderation Model
The second set of analyses examined whether the aforementioned variables moderated the relationship between ACEs and the harm caused by a mass public shooting. Table 2 shows the results of the linear regression models testing the interaction effects. Of the six potential moderating variables examined, three emerged as significant predictors including signs of a personal crisis, having a mental health diagnosis, and the number of recent stressors.
Linear Regression Models Testing the Interaction Effects Predicting the Number of Victims in Public Mass Shootings.
p < .05, ** p < .001.
The results of these models show that the interactions between ACEs and signs of crisis, mental health diagnosis, and the number of stressors are all significant predictors of the total number of victims. Each of the interaction models explains a small amount of the variance in the number of victims (Column 5).
To illustrate the relationship between ACEs and harm moderated by signs of a crisis, Figure 4 graphs this relationship for a hypothetical shooter with an average (
Interaction Between ACEs and Signs of Crisis Predicting Total Number of Victims.
Figure 5 shows the interaction between ACEs and having a mental health diagnosis in predicting the number of victims in a mass shooting. For those shooters with a mental health diagnosis, there is no relationship between the total number of ACEs and the total number of victims. For shooters without a mental health diagnosis, there is a strong and positive relationship between the total number of ACEs and the total number of victims.
Interaction Between ACEs and a Mental Health Diagnosis Predicting the Total Number of Victims.
The interaction between the number of stressors and ACEs is depicted in Figure 6. Figure 6 graphs this relationship for a hypothetical shooter with an average (
Interaction Between ACEs and the Number of Stressors Predicting the Total Number of Victims.
Limitations
By examining mass public shootings and using a more restrictive definition (e.g., at least four victims killed by firearms, some of the offense must take place in public, excluding events with criminal motivations) than other researchers such as those using the gun violence archive, we are only examining a subset of the larger category of “mass shootings.” However, this also provides the opportunity to dive deeper into the background characteristics of this smaller number of shooters, offering us a unique opportunity to examine childhood trauma and similar issues.
The validity of some of the measures is questionable; as a good portion of the shooters die at the scene, most of the information is collected from retrospective investigations into these offenders’ lives and relies heavily on publicly available materials. However, it should be noted that the investigations for each of the shooters were coded independently by two individuals and discrepancies were identified and analyzed (Peterson & Densley, 2022). Likewise, we were unable to examine the influence of a variety of aggregate-level measures that might influence mass public shootings, such as gun legislation, cultural influences, and community-level risk factors. Finally, since key variables in the data did not include temporal information, such as dates, we were unable to thoroughly examine the sequencing of these events.
We have no doubt that recent efforts to expand the definitions of ACEs including those who advocate adding additional individual, household and community-level adverse experiences are worthy of investigation (Afifi et al., 2020; Finkelhor et al., 2015; Wade et al., 2016). Although our research was limited by the use of secondary data, future researchers should continue to investigate the myriad of various individual experiences both at the family and community that contribute to the complex issue of childhood trauma that cumulatively leads to a variety of negative outcomes including an increase in violent behaviors.
Implications for Social Work
This study sought to understand if ACEs and other risk factors predict the number of victims in mass public shootings involving firearms. Specifically, we examined whether ACEs had a direct, indirect, or moderated impact on the number of victims. Using data from the years 1966–2022 from the Violence Project, our results provide compelling evidence to counter prevailing public perceptions of who may engage in a mass public shooting. Although we did not find ACEs to have a direct impact on the harm caused by a mass shooting, we did find evidence that the influence of ACEs was both mediated and moderated by several other factors previously associated with mass shooting offenders and the number of victims in such events.
In examining the mediated effect of ACEs, we find that ACEs have an indirect effect on the number of victims through their influence on fame-seeking rationalizations. Although the examination of individual ACEs is beyond the scope of this project, we did conduct supplementary analysis to investigate the impact of several individual ACEs shown to be important predictors of our variables of interest in a bivariate analysis. The results of the models including those individual predictors were identical to the results of the reported analysis; the only influence of the individual ACEs examined was through fame-seeking motivations.
This is especially important for social workers as previous research has shown that shooters tend to disclose their intentions ahead of time (Freitas & Annas, 2022; Lankford et al., 2019). This is known as “leakage,” meaning that others are informed of the shooter’s intent, prior to the shooting event (Lankford et al., 2019; Lankford & Silva, 2021). A key implication from our work is that social workers who interact with individuals exhibiting signs of fame-seeking or the desire for notoriety may want to explore if this is a warning sign or cue that these individuals may pose a risk to themselves or others. Our findings indicate that fame-seeking is more likely to signal interest or intent to pursue this type of gun violence, rather than experiencing multiple ACEs or having a mental health diagnosis. However, further research is needed to understand if and how fame-seeking desires develop as a possible response to experiencing one or more ACEs and to which ACEs.
Along with this finding, we also found that the presence of multiple firearms at the shooting scene had a direct relationship to the amount of death and injury caused during the shooting event. However, it is important to note that most of these weapons were obtained legally and that often, these firearms were collected specifically to complete a planned mass public shooting event (Silver et al., 2018). In fact, Lankford and Silva (2021) found that the median time offenders acquired their first firearm before the attack was 7.8 months and that the last firearm was typically purchased 1.2 months before the attack.
These two findings suggest that it is particularly important for social workers to follow the lead of physicians in screening for the presence of firearms in their client’s lives and discussing gun safety with their clients (Roszko et al., 2016; Slovak et al., 2008). With proper preparation and training, social workers can engage in conversations about gun ownership, use of, and safety practices including safe storage in the home, with clients who either disclose gun ownership or through an assessment process, such as those initiated by medical personnel during routine visits (Richards et al., 2022; Wintemute et al., 2016). Our results suggest that this would be particularly salient for clients exhibiting narcissistic or fame-seeking tendencies.
Furthermore, for individuals who exhibit these fame-seeking characteristics and who possess or potentially have access to firearms, Extreme Risk Protection Orders (ERPOs) or what is commonly known as “red flag laws” (Cornick et al., 2023) would provide a mechanism to address concerns for public safety. These laws allow for family members and law enforcement to remove any firearms and ammunition in the possession of an individual who is at imminent risk of harm to themself or others for a certain period, usually 6 months to a year (Barnhorst & Rozel, 2021). A family member of an individual may be a key source for this information. However, social workers should be aware of issues of client’s confidentiality regarding engaging in any outreach to others regarding these items (Cornick et al., 2023).
The results from our moderation analysis suggest that while the presence of a mental health diagnosis, the number of stressors a shooter is experiencing, and the signs of crisis experienced by the shooter, all moderate the relationship between ACEs and the number of victims, it is in a way that is counterintuitive to the public’s perceptions of mass shooters. We find that shooters without a mental health diagnosis, and those with fewer signs of crisis and ongoing stressors cause the most harm in terms of the number of victims.
Yet, media portrayals and legislators (Hirschtritt & Binder, 2018) continue to foster public perceptions of dangerousness and a desire to distance themselves from those who experience ACEs and mental health challenges (Purtle et al., 2022). Social work can help to counter these viewpoints (Sperlich et al., 2022). Efforts can be as simple as engaging in individual advocacy efforts to refute these false narratives to lobbying for local, state, and federal legislative action to address gun violence (Simonsson & Solomon, 2021). Not surprisingly, the profession of social work, including our national organization, is being asked to do more to address the national crisis of gun violence which is affecting every system where we provide services (Bright, 2022; Dorris & Murphy, 2023; Logan-Greene et al., 2018).
Even though not related to the amount of harm caused in the event, most of the shooters in our sample were adult males. As a group, adult males are often hesitant to seek assistance for mental health issues and as adults, these individuals cannot be forced into treatment unless they have reached the stage where they are a clear danger to themselves or others (see Seidler et al., 2018 for a review of the issues in engaging men in psychological treatments). This makes preventive treatment more challenging with this population. However, as mentioned above, with proper preparation and training, social workers can assess firearm violence risk and collaborate with other professionals to address warning signs.
Conclusion
Our research suggests that we need to move away from the narrative that mass public shootings are carried out by those who experience ACEs or are mentally ill individuals who have reached a breaking point and impulsively engage in mass violence. Other scholars have called for ending this false association as well (Bright, 2022; Dorris & Murphy, 2023; Simonsson & Solomon, 2021). Indeed, our findings show that individuals with traumatic childhood experiences who exhibit signs of personal crisis, have a diagnosed mental illness, and/or are experiencing stressors cause less harm than those who do not have these risk factors. Social work can tackle the complicated issue of gun violence by employing multifaceted approaches to help refute false narratives of who may perpetrate gun violence and enhance our prevention strategies by learning more ways to assess for firearm violence risk within our practice areas.
Footnotes
Disposition editor: Cristina Mogro-Wilson
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, authorship, and/or publication of this article.
