Exploring hazing experiences and perceived physical and mental health outcomes in the United States military through a cross-sectional study

Abstract

Background

Hazing–defined as an act that is required to gain entry into a group and humiliates, degrades, abuses, endangers, or otherwise embarrasses an individual, irrespective of willingness of the person–is a persistent problem in the United States (US) military, and can manifest in physical and emotional harm, and in extreme circumstances, death. However, the presentation of hazing, including the exact experiences and resulting outcomes is scarcely studied, leaving a gap in understanding and a need to specifically investigate this issue.

Methods

A cross-sectional study was conducted by administering a survey with questions about hazing experiences and outcomes to 227 service members at a military installation in the Southeast United States over 14 months. Frequency of hazing experiences and outcomes were derived. To understand how the large number of variables related to each other, a cluster analysis provided Jaccard measures of similarities amongst experience variables. Logistic regressions were conducted to test the association between the experiences and outcomes of hazing.

Results

A total of 227 service members answered the survey; the final cleaned sample resulted in 202 participants who were primarily active-duty (94.0%), in the Army (95.6%), enlisted rank (E1–E4; 58.5%), male (69.6%), and white (49.4%). Almost one-fifth (19.8%) of the participants endorsed experiencing hazing at the installation, and over a fifth endorsed experiencing hazing in the form of sexual violence (22.4%), being tricked (27.8%), and acting like a servant (22.3%). Hazing experiences were significantly associated (p < 0.05) with greater odds of several types of negative physical (ORs ranging from 2.8 to 15.9) and mental health outcomes (ORs ranging from 4.7 to 68.0) and multiple effects on the organization, such as desire to leave the military (ORs ranging from 2.9 to 34.1), since arriving at the current military location.

Conclusions

Hazing manifests in various manners and results in harmful consequences for both individuals and the organization. The present study provides an understanding of hazing through descriptive epidemiology, which can help both leaders and practitioners address the issue.

Background

Hazing–defined as an act that is required to gain entry into a group that humiliates, degrades, abuses, endangers, or otherwise embarrasses an individual, irrespective of willingness of the person [2]–is a pervasive problem in the United States (U.S.) military. Although this shared definition of hazing applies across multiple entities (eg, sports teams, fraternities, and military units), it is also worthwhile to specify the definition of this issue in the organization studied. In the context of the Department of War, (DOW [45]), it is defined as:

“A form of harassment that includes conduct through which Service members or DoD employees, without proper military or other governmental purpose, but with a nexus to military service, physically or psychologically injure or create a risk of physical or psychological injury to Service members for the purpose of initiation into, admission into, affiliation with, change in status or position within, or continued membership in any military DoD civil organization.”

The policy goes on to state that hazing can be committed electronically and via social media and includes (but is not limited to): striking or threatening to strike someone, pressing an object onto one’s skin; berating to belittle or humiliate; pressuring someone to do illegal, harmful, or demeaning acts; playing cruel or abusive pranks; marking, restraining, or altering someone’s body; using water on someone in an abusive way; or forcing someone to eat, drink, or take substances. Finally, it is indicated that acts necessary for governmental purposes and properly directed by Command are not included in these acts.

Hazing incidence in the Department of War

Between 2017 and 2020, approximately 200 hazing cases were officially reported in the U.S. military. However, confidential self-reported surveys suggest that thousands of incidents occurred when taking a more comprehensive and thorough approach to defining hazing to include harassment and sexual violence, indicating that current mechanisms do not truly track incidence [48]. Moreover, in the Millennium Cohort Study by the DOW, a survey of over 250,000 active-duty, reserve, and national guard members followed over decades, 10% of participants reported experiencing hazing in a military context [9].

Coinciding with the timing of this study, the RAND corporation developed a survey to examine hazing and bullying in the forces at the direction of the DOW in response to Congressional direction [28]. This developed survey included questions that prompted for: (1) threatening or causing physical pain; (2) deprivation of food, water, or sleep outside of official reasons; (3) harmful consumption of food, water, or alcohol; (4) cruel pranking or abuse not for military training (eg, being taped); (5) taking or damaging belongings; (6) pressuring to commit illegal acts (eg, steal something); (7) conducting acts to humiliate (eg, spreading false gossip online); (8) upsetting insults (eg, name-calling); (9) encouraging engagement in disgusting acts (eg, drinking gross water); and (10) touching private parts [28]. References that may result in classification of sexual assault (such as item 10 or name-calling that is sexual harassment such as calling someone the slur “fag’”) were removed, resulting in the final estimates of hazing employed in a survey by the 2017 Workplace and Equal Opportunity Survey of Active-Duty Members [33].

Results of this survey included an overall estimated prevalence of hazing at only 3.5% of the DOW [34]. Of note, however, harassment (to include sexual and racial) can also be considered hazing if done as an attempt to initiate someone into a group and aimed at humiliating them, and sexual violence and hazing can co-occur [28, 33, 48]. Thus, only focusing on this metric (3.5%) is limiting in the scope of understanding. In such circumstances, of those who experienced racial/ethnic harassment in the DOW, 6% of personnel described incidents as hazing [35], 15% of women and 23% men in the DOW defined their “worst” incident of sexual harassment as hazing, and approximately 9% of women and 30% of men described “worst” situations of sexual assault as hazing in the Army [32, 33]. The work by RAND and subject matter experts–along with direction by Congress–indicates that these incidents should also be under the umbrella of hazing, thus the 3.5% estimate is not a comprehensive metric to fully understand hazing in the forces as it is limited by current policy and tracking mechanisms. Furthermore, since these hazing surveys are a part of wider efforts and Service Members (SMs) are limited in time and often logistical concerns (eg, internet access), these metrics employed across the DOW are not as inclusive of the detail the authors employed for examining this act (eg, naming specific types of physical pain, assessing military-specific acts, or specifying humiliating acts). Accurately determining the incidence of hazing is challenging due to underreporting, driven by SMs inability to clearly identify and define the acts as problematic and reflective of this interpersonal violence, a culture of silence, and the general acceptance of such behavior [39].

Outcomes of hazing

Hazing in the civilian sector is associated with numerous psychological and physical consequences for those who experience this violence, including severe pain; symptoms of depression, anxiety, and posttraumatic stress disorder; and in extreme cases, death [2, 14, 18, 25, 38]. Hazing can affect relationships with others, belief in leaders, and erode trust and morale in the organization [23, 49]. Prior research suggests that those who experience hazing feel betrayed by the organization and lose interest in it [49]. In the military, this could also lead to degradation of comradery, lack of cohesion, and distraction from the mission. Although hazing has been noted as an issue within the instiution for over a century [27], the literature in the U.S. military sector is nascent as the research to date is limited in scope.

Current studies lack comprehensive overview of all hazing behaviors [17, 37], do not differentiate between hazing and bullying experiences [36], are exclusive to a deployed environment [7], or are focused on suicidality [12, 24]. The DOW is moving towards more comprehensive tracking of hazing incidents [34], however, a comprehensive presentation of types of hazing experiences in the U.S. military and resulting outcomes is not currently documented. This gap in understanding is dire as fatal outcomes from these experiences continue to occur. Illustrative of this detrimental outcome is the dozens of U.S. SMs who have died by suicide following hazing incidents over the past decade [29].

Examining hazing in the D epartment of War

Hazing in the U.S. military is nuanced. SMs undergo rigorous physical and mental training, particularly as new recruits [26]. During training, the line between acceptable behavior and reportable behavior could quickly blur [37]. New recruits are often more loyal to their immediate social group than to the military or the mission; this can create an atmosphere in which groups act together to haze and/or those hazed do not report it as they perceive it as positively becoming included in the group [41, 42]. Nevertheless, there is a clear distinction between these acts that are necessary to prepare for battle and those that are done to humiliate another for initiation into the group.

A Congressional Research Service report suggests that adapting programs from the civilian sector—such as those used on college campuses—could be beneficial for addressing hazing within the U.S. military [21]. Moreover, there is an identified need to build capacity, articulate prevention methods, create formalized consistent reporting and prevention efforts, and reduce risks associated with hazing [48]. To current knowledge, prevention efforts include training for leaders that defines hazing, discusses misconceptions around hazing, and provides information on formally reporting and prosecuting hazing [23]. The next step in this prevention approach is to understand how hazing manifests and its outcomes within the U.S. military as this will equip the organization with information needed to address the issue. Without these data, prevention programs cannot be effectively fitted to incorporate the specific needs of the U.S. military while reducing these behaviors.

Therefore, this study aims to document hazing experiences and outcomes at a U.S. military installation through a descriptive, cross-sectional study. Specifically, we investigated the proportion of individuals who endorsed experiencing military specific [eg, blood-pinning [31]:slamming a pin into one’s chest after Airborne qualification], nonphysical, violent (inclusive of sexual violence), or other physical hazing acts, and we examined how those experiences related to cohesion (feeling like part of the group, trouble at work, or desire to get out of the military), mental anguish (feelings of depression, stress, humiliation, being in danger, and difficulty in relationships), and physical stress (injuries and difficulty sleeping). As research is scant in the field, this study provides a rich source of information from a novel population and is hypothesis generating rather than testing.

Methods

A cross-sectional study was conducted to investigate the experiences and outcomes of hazing by distributing an online survey on hazing experiences and outcomes at a large military installation in the Southeastern U.S., a location with over 52,000 military personnel. The study team regularly reviewed demographic data at intervals to ensure racial, ethnic, and gender representation of the installation. (If these standards were not met, the team would have employed targeted recruitment). Data were collected during 2021–2023 from a convenience sample of 227 SMs. The study team recruited SMs over the age of 18 through advertisements (eg, posted flyers) and in-person outreach for a peer and group experiences survey. Participants completed the online via a commercially available software (either via a QR code or on a provided electronic device). Participants were assured that the survey was anonymous, with no linkage to the participant in the future. Participants provided electronic consent; in accordance with DOW regulations, SMs self-endorsed that they were “off-duty” (not actively participating in military activities), and thus, were eligible to receive a $10 USD gift card (pre-loaded individual giftcards were provided) as compensation for participating in the research. The survey took approximately 15 min to complete. Participants were able to endorse all acts that they experienced (ie, they were not required to only choose one). Inclusion criteria consisted of: (1) active-duty status (active duty, activated reservist, or national guardsmen); (2) at least 18 years of age; (3) able to speak and comprehend English sufficiently to complete study instruments; and (4) “off-duty” delineation as defined above. The study was approved by the local Research Protections Office.

Due to the absence of validated surveys specifically designed to assess hazing within military settings at the time of initiation of this study, we adapted established instruments from previous research on hazing and violence among college students [1–3, 20]. This previous survey has been implemented in numerous civilian populations and has been found to be reliable and valid. This modified version resulted in a survey that captures SMs’ experiences and outcomes regarding hazing. To ensure participants understood we were referring to hazing experiences, we started questions by stating “Feeling like a part of the group is important. Please click one if you were encouraged to do any of the following to be accepted by your peers since arriving at [military installation].” Hazing occurs as an initiation to a group, and thus, this marker in time is most appropriate as compared to a stagnant metric (eg, 12 months) that may miss this pivotal moment. Although it could be nuanced to discriminate between tasks that were required for job duties and those that constitute as hazing, this prompt for these questions stated to participants that these actions should be considered as those required to gain entry to a group provides a delineation. Other research efforts have included survey creation and validation since the inception of this study [28], although the authors were unable to use these other surveys, the survey utilized has been tested and refined in tens of thousands of college students and is the most widely used instrument to date to assess hazing prevalence. Moreover, in line with the instrument utilized in this study, other instruments also capture similar constructs (eg, physical or sexual acts) surrounding the intention to haze as an act of humiliation for entry to a group and, specific to this population, delineate the difference between acts required for job duties [7, 28].

The research team first adapted the survey from the civilian sector and drafted it for military applicability. Then, following a pilot effort–which included 10 SMs and 10 civilian/contract personnel, prompting for clarity and understandability–and subsequent question refinement, primarily including alteration of language and typos, deleting questions that were obtuse and difficult to understand the applicability in this population, the final survey comprised 21 items aimed at evaluating the hazing experiences and outcomes experienced by SMs. Participants were asked about experiencing acts of hazing that included physical or sexual violence, military specific acts (eg, blood-pinning), non-physical acts (eg, being forced to act as a servant), physical non-violent acts (eg, being forced to eat gross food), and psychological, physical, and organizational outcomes. These metrics also align with DOW defined hazing behaviors with the addition of behaviors under this category that civilians subject matter experts on hazing would define as part of the behavior [28]. Aligned with previous work, participants were first asked about experiences that have previously been found to occur more commonly, followed by those that are less common [18]. A full list of questions can be found in Table 1. Of note, these efforts employed since the inception of the study, do not provide the same granularity as this research study [28]. Questions were comprehensive in an effort to better understand this understudied experience and its consequences at the location.

Table 1.

Select survey questions

Category	Question
Activities	Have you ever been hazed to be accepted as a member in a [military installation] unit?
Feeling like a part of the group is important. Please click one if you were encouraged to do any of the following to be accepted by your peers since arriving at [military installation]
Non-physical hazing	Act as a personal servant to others
	Ordered to perform a task that would trick you into getting in trouble or humiliate you
Military specific hazing	Hang upside down from a tree or pole (“koalafying”)
	Blood-pinning (ie, bloodwings)
Non-violent physical hazing	Drink or eat gross stuff (ie, bodily fluids, combinations of food/vile substances, etc.)
	Drink large amounts of an alcoholic beverage
	Be tied up, taped, or confined to small space
	Secure items to yourself using string or rope
Physical violence as hazing	Be punched or slapped in celebration of an event (i.e., "pink belly," promotion)
	Be whipped, kicked, or beaten
	Participate in games where you were required to hit or slap someone or be hit or slapped yourself
Sexual violence as hazing	Simulate sex acts in front of the same gender
	Simulate sex acts in front of other gender
	Do sex acts with the same gender
	Do sex acts with another gender
	Display sexual body parts in front of a group or individual
	Be nude or partially nude in front of the group or in public place
As a result of participating in any of these activities, did you:
	Feel less like a part of the team or group?
	Feel humiliated or degraded?
	Feel ashamed, guilty or depressed?
	Feel in danger?
	Feel stressed?
	Have difficulty sleeping?
	Incur physical injuries?
	Have trouble with work?
	Have problems in relationships with friends and/or family?
	Consider getting out of the military?

A multiple part question classified as sexual violence at the installation:

During your military service, has someone ever…(harassed you… in a way that made you feel unsafe, taken sexual photos… when you didn’t want to, kissed you…when you didn’t want to, fondled or grabbed your sexual body parts, removed or attempted to remove your clothing, made sexual comments, jokes gestures or looks about/to you, said you were gay or lesbian as an insult)?…

During your military service, when you were drunk, high, drugged, or passed out and unable to consent, have people ever ...(had vaginal sex, made you perform/receive anal sex, make you perform/receive oral sex)?…

Considering the definition of hazing provided in this survey, did any of the previous items occur as a form of hazing within a [military installation] unit?

Collected demographic characteristics included race/ethnicity, branch, active-duty status, rank, and gender; characteristics of the study population are presented in Table 2. Heterogeneity in the population sampled could affect hazing behaviors and outcomes experienced. There was not, however, sufficient sample size to conduct analyses by gender and/or rank.

Table 2.

Characteristics of study participants

Characteristic	N = 202	% (Total 100)
Military status
Active-duty	169	94.0
Reserves	5	2.8
Other	6	3.3
Branch
Army	172	95.6
Air Force	2	1.1
Marines	3	1.7
Navy	3	1.7
Rank
E1-4	103	58.5
E5-6	44	25.0
E7-9	7	4.0
O1-3	16	9.1
O4 or higher	4	2.3
W1-4	2	1.1
Gender
Male	120	69.6
Female	52	29.7
Other	3	1.7
Race
White	85	49.4
Hispanic/Latino	32	18.6
Black	28	16.3
Asian	14	8.1
Other	7	4.1
Native American	3	1.7
Native hawaiian or other Pacific Islander	3	1.7

Due to rounding, percentages may not equal 100

Data analysis

Data were cleaned to exclude participants who failed to respond to any questions about hazing experience, resulting in a final sample of 202 participants (from the original 227 SM sample). The frequency of hazing experiences and outcomes were summarized using descriptive statistics. To explore whether individuals who experienced a given hazing experience also experienced other specific hazing experiences, a cluster analysis was conducted by calculating Jaccard similarity measures–a simple, normalized score–among experience variables and reporting clustering of variables with a dendrogram [19].

There were ten outcomes of interest. Outcomes were simply dichotomous in “yes” or “no” answers. For each outcome, we used logistic regression to identify which, if any hazing experiences were associated with the respective outcome. Univariate regression models were used to consider each hazing experience as an independent variable in isolation (ie, unadjusted). In doing so, the goal was to examine these relationships at high-level due to the exploratory nature of this study. Given the many limitations of potentially overcorrecting during adjusted analyses [13], analyses are presented as unadjusted. Analyses were conducted with consideration for rank and gender as covariates, however, due to the limited sample size, these were not viable for logistic regressions. Analyses were performed using SAS 9.3M2 (Cary, NC), and Stata 18 (College Station, TX).

Results

Hazing experiences

Figure 1 reports the prevalence of hazing acts experienced. When asked if they experienced hazing at the installation, 19.9% of the sample endorsed “yes.” When asked if the participants experienced sexual hazing at the installation broadly, 22.4% of the sample endorsed “yes.” When asked specific questions about sexual violence, 9.8% of participants reported being forced to engage in sexual acts with the other gender. Other sexual violence acts (eg, sexual acts of the same gender) were the least endorsed experiences with the lowest prevalence (2.6–7.0%). Additionally, 22.3% of participants endorsed being forced to act as a servant and 27.8% reported being tricked. Regarding military specific acts, 15.3% of participants endorsed experiencing blood-pinning and 11.6% of participants endorsed experiencing koalafying (ie, being forced to hang upside down from a tree).

Fig. 1.

Endorsement of experiencing hazing behavior or outcome (N = 202)

Cluster analysis revealed two distinct profiles of hazing experiences and characteristics of each. As illustrated in Fig. 2, experiences and activities linked by shorter branches (closer to 0 as indicated by the length of the line) co-occurred more—meaning participants who reported one experience were also likely to endorse experiencing the other experience—compared to those connected by longer branches (as indicated by longer lines).

Fig. 2.

Dendrogram of hazing behaviors

In the first broad profile, being forced to eat gross food commonly co-occurred with being confined. These experiences also frequently co-occurred with being hit during a game and experiencing "hazing" (defined as being hazed on the installation). Within this profile, being forced to be nude was strongly associated with blood-pinning, and both were often linked to being forced to drink alcohol. Participants who reported being tricked also reported being hit or slapped in celebration of a promotion and frequently experienced hazing as sexual violence. Additionally, those forced to act as a servant were often required to secure an item to themselves, while those encouraged to participate in sex acts with individuals of a different gender were also required to display body parts.

The second broad profile was characterized by experiences involving sexual violence and humiliation. For instance, koalafying often co-occurred with being encouraged to simulate sexual acts with someone of the same gender. These experiences were further associated with being whipped, which in turn co-occurred with being encouraged to simulate sexual acts with individuals of a different gender and engaging in sex acts with those of the same gender.

Hazing outcomes

Participants most commonly endorsed feeling stressed (51.1%), desire to get out of the military (47.2%), and difficulty sleeping (38.6%). Participants least commonly endorsed feeling in danger (11.4%). Prevalence of endorsement of other outcomes ranged between 25.6%−36.5%. All outcomes and their prevalence can be found in Fig. 1.

Figure 3 reflects odds of desire to get out of the military, difficulty sleeping, and feeling stressed, respectively, by hazing experience. Count reflects the number of participants that answered questions both regarding the experience and the outcome. These outcome variables were chosen as they had the greatest prevalence of endorsement by participants.

Fig. 3.

a Odds of desire to get out of military by hazing experience. b Odds of difficulty sleeping by hazing experience. c Odds of feeling stressed by hazing experience

All hazing experiences were associated with greater odds of desire to get out of the military (2.8–34.1; p < 0.05). Participants who experienced sexual violence as hazing were at 34.1 (95% CI 7.7–151.0) times the odds of endorsing desire to get out of the military as compared to participants that did not experience sexual violence as hazing. Participants who experienced specific acts of hazing as sexual violence that occurred with another person (eg, simulating/participating in sexual acts, displaying body parts), were at over 8 times the odds (8.6–19.0, p < 0.05) of endorsing desire to get out of the military as compared to those who did not experience these behaviors. Of the experiences analyzed, those who reported being whipped/kicked/beaten as a hazing behavior were at the lowest odds of reporting a desire to leave the military comparatively to other behaviors; however, this finding was not significant (2.7; 95% CI 0.7–11.0). Those who experienced military specific acts (ie, blood-pinning and koalafying) were at 3.5–7.2 (p < 0.05) times the odds of desire to get out of the military.

Most hazing experiences were associated with greater odds of difficulty sleeping. Participants forced to act like a servant were at 16.0 (95% CI 6.5–39.0) times the odds of reporting difficulty sleeping than those who did not report being required to act like a servant. Those who experienced military specific acts (ie, blood-pinning and koalafying) were at 2.8–4.9 (p < 0.05) times the odds of difficulty sleeping. Participants who experienced sexual violence as hazing were at 12.7 (95% CI 5.0–32.2) times the odds of endorsing difficulty sleeping as compared to those who did not experience sexual violence as hazing. Being forced to be nude and participate in sex acts with the same gender were not significant in determining odds of difficulty sleeping (2.3, 95% CI 0.8–7.1; 6.8, 95% CI 0.8–62.4).

Many hazing experiences were associated with greater odds of feelings stressed. Participants who experienced sexual violence as hazing were at 68.0 (95% CI 7.7–515.0) times the odds of endorsing feeling stressed as compared to participants that did not experience sexual violence as hazing. Those who experienced the military specific acts of blood-pinning and koalafying were at 2.3 (95% CI 0.9–5.4) times and 9.9 (95% CI 2.2–44.5) times the odds of feeling stressed, respectively. Participants were at 34.1 (95% CI 7.9–147.6) times the odds of feeling stressed if they were required to act like a servant. Hazing that occurred as being forced to have sex with the same gender and simulate sex acts with the other gender were not significant in determining odds of feeling stressed (4.1, 95% CI 0.5–37.4; 5.2, 95% CI 0.6–45.3). Experiences not presented in forest plots (eg, sex acts with same gender in Fig. 3c) did not have enough participants endorse the experience for analysis.

Discussion

There are several notable findings in this study including the frequency of experiencing hazing, co-occurrence of experiencing violent behaviors, increased odds of physical and mental health outcomes associated with experiencing hazing, and negative consequences of hazing that are unique to the U.S. military.

Presentation of hazing experiences

In previous work by the DOW, hazing was estimated to be at 3.5% of the force, which was not inclusive of the metrics of racial/ethnic harassment, sexual harassment, or sexual assault cases [32, 33]. In this study, when participants were asked directly if they experienced hazing at the installation, less than 20% endorsed experiencing hazing, however, when asked in this survey about specific acts of hazing (eg, acting as a servant), participants reported greater prevalence (> 20%). This suggests participants may not consider or be able to delineate all acts of hazing as experiencing hazing and are not always able to identify the behavior. Previous literature in military populations suggests that groups have a difficult time defining hazing, even when provided training on the topic [39]; this difficulty is especially prominent when distinguishing between hazing and bullying. These data further highlight that it is difficult for SMs to define and identify hazing appropriately. Understanding that the behavior is problematic and labeling it as violence–with knowledge it should be reported as such–is important in understanding prevalence, accountability, and addressing this issue.

Although different in many ways (bound by contract, a lack of autonomy over location and schedule, and varied levels of responsibility), collegiate students are a parallel population to the U.S. military in age group, the shared locality of work and living spaces, training for occupation, and similarity in risk taking behavior. In the college setting, Allan [4] found that 47% of students reported experiencing hazing behaviors, with almost 75% of specific groups (sport and fraternities) experiencing hazing. The most common types of hazing in these groups includes alcohol related experiences (eg, forcing to drink large amounts of alcohol), and being forced to chant or sing in public [4]. Almost 20% of participants in this study at a U.S. military installation also reported being forced to drink alcohol, which was one of the more commonly reported experiences. In contrast, experiencing sexual violence and being forced to act as a servant presented as common hazing behaviors experienced in this study. This indicates that while some similarities exist between the populations, hazing may present differently in different ways. Thus, prevention programming may need to be specifically tailored to the U.S. military to be successful.

More participants reported experiencing sexual violence as hazing than those that reported hazing generally when directly asked. There is no literature, to our current knowledge, that specifically addresses sexual violence occurring as hazing in the U.S. military; rather, reports/publications typically addressing hazing broadly with an acknowledgment that sexual violence is a type of hazing behavior or indication of hazing as a form of sexual violence. Nevertheless, Congress has tasked the DOW with addressing hazing incidents of sexual violence [47] and the DOD plan states that sexual violence training should include information regarding the intersection between hazing and sexual violence [46]. According to recent DOW reports, 15% of women and 23% men in the DOW defined their worst incident of sexual harassment as hazing and 9% of women and 30% of men described their worse situation of sexual assault as hazing in the Army [32]. Sexual violence that occurs as hazing also involved more perpetrators compared to assaults that were not indicated as hazing [30]. Behaviors that were identified as sexual violence under the guise of hazing tended to co-occur in this current study, align with these previous reports. Taken together, this work further supports that the co-occurrence of sexual violence and hazing warrants further investigation.

Of note, we are unable to distinguish the reasoning why many behaviors (eg, blood-pinning and sexual violence) co-occurred together in this quantitative study; further qualitative work would be necessary to further distill this relationship. Further, rates of sexual violence in the military are currently estimated at less than 10% [32], this discrepancy could be due to the specific population surveyed.

Outcomes of experiencing hazing behaviors

Common outcomes of hazing in this study included feeling ashamed and stressed, which is aligned with previous research that found being hazed or bullied on a deployment was associated with negative mental health outcomes such as major depressive disorder and post-traumatic stress disorder [7]. One of the most reported outcomes in this study was also difficulty sleeping. In the U.S. military, less than two-thirds of SMs get recommended sleep [15], which can have immediate consequences such as fatigue and reaction time [16] to long-term consequences such as elevated risk of cancer and neurodegenerative diseases [43, 51]. This current research indicates that experiencing hazing behavior is a risk factor for experiencing difficulty sleeping which can then lead to negative mental and physical health outcomes (eg, feeling stressed or incurring physical injuries), resulting in outcomes that are especially compounding.

Hazing acts are presented as a requirement to enter a group [44], and rooted in misconceptions around this violence contributing to group cohesion [8, 22], solidarity [10, 11], and identity [25, 49, 50]. Despite these fallacies, however, hazing leads to a degradation of relationships, lack of trust in the organization, decrease of morale, and perceptions of betrayal of fellow members and the institute [49]. In the current study, all hazing experiences were associated with greater odds of desire to get out of the military (ORs: 3–34). Moreover, participants were also at increased odds of experiencing trouble at work, further showcasing the effect of this negative experience on productivity and work environment. Overall, hazing had a negative effect on perceptions towards the institution, leading to a lack of retention. This negative effect is aligned with previous research.

The U.S. military also prides itself on its community, a cornerstone of the structure [6] , as indicated by regular events and general cohesiveness of the services. Those who experienced hazing were also at greater odds of having problems in relationships with friends and family. The military can be an organization with strong internal community as co-workers become friends, and families of SMs regularly interact with each other. Further, many SMs have peers of similar age that they regularly work with and are in isolating conditions from others (eg, locality or working hours), which can quickly lead to close bonding with co-workers. If SMs experience hazing behavior at work, it is not surprising that they would have difficulty with friendships –as these friends are likely also coworkers that could be aware of the hazing acts–, which can further degrade the morale and unit cohesion, as personnel feel unsupported in their work environment since their friends and coworkers are often the same population [49]. In fact, in a subset of data from this study, the authors found that less than half of SMs that experienced hazing reported it, which can further lead to isolation and feelings of belonging and support [5]. Thus, this study further illustrates the detrimental effects of this behavior on interpersonal relationships.

Understanding this behavior and the consequences can inform prevention programming, which would need to be tailored to the community it is addressing (eg, in this work the population was primarily young, male SMs in infantry roles).

Limitations

Although this work provides vital information, there are limitations. The reported odds ratios also had large confidence intervals, showing a lack of precision. Some of the results that were not significant may be due to a small sample size; similarly, a larger sample could result in smaller p value for those that were significant with more precise confidence intervals. Due to the small and heterogenous sample, the study may have failed to detect relationships that are meaningful but did not display statistical significance.

Behaviors and experiences that are listed are broad and encompassing but do not include all questions that one may classify as hazing in the civilian sector, such as sleep deprivation or requiring physical activity. This highlights the difficultly of studying this behavior in the U.S. military. Also, as we created a scale specifically for this work–since there was not one currently available at the time of study inception–it is possible there was increased subjectivity in the prompts asked. Some questions may be more prone to interpretation such as those questions referencing “body parts.” It is also possible that some acts were endorsed but a part of typical military training [eg, one could have said “yes” to being forced to eat gross food in reference to Meal, Ready-to-Eat (MREs) as the only available food]. It was, however, indicated that these actions were required to gain entry to the group as opposed to acts that all may be experiencing as a part of training for the military mission. Future work could further indicate, such as in the survey developed by the RAND foundation [28], that these acts were not done as part of official military duties. Metrics for hazing should align with the context in which they are studied (eg, in this setting there is a need to be military-specific); some may note that the authors more broadly included behaviors of hazing than currently included in DOW policy. Overall, the scale was based on prior work for decades in the civilian sector, however, this context varies, as may the interpretation of the questions.

The data that were cleaned due to non-response could have included participants that were lower in rank, and younger in age, due to fear of reporting. In other acts of interpersonal violence (eg, sexual violence), it is reported that this group (lower in rank and younger in age) is least likely to report experiencing these acts but would be likely to experience negative outcomes [40]. This would result in bias towards the null and make it smaller in absolute value, with our conclusions being underrepresented in magnitude. Finally, it is possible that participants enrolled that were more comfortable taking the survey as they did not experience the behaviors or outcomes. In contrast, this would yield a bias would away from the null.

We were unable to analyze by rank, race, and gender due to this limited sample size. As this work was exploratory in nature, we also did not collect or analyze data that could affect these experiences such as age, time in service, deployment, or specific training experiences. It is possible that specific trainings could confound the relationship observed between the exposure and outcome variables. Future research should examine these factors as experiences may vary by these demographics, as indicated in other work to date [34].

Although the survey was administered to over 200 participants, not all participants answered questions and this survey took place at an installation with a unique military population as the location has mainly combat, infantry Soldiers,–hazing experiences in this particular military occupational specialty may vary, since infantry SMs are often in the field as a unit to prepare for combat–and as such, our ability to generalize is limited. These data are also specific to and contextually relevant to the location where and the time in which it was collected. These data can be considered relatively small in the context of the larger DOW. Timing of data collection is based on a myriad of factors in this environment to include troop movement and fluctuation in available population to sample. Of note, the world was also experiencing a global pandemic during this time, and unique attributes of that, such as increased isolation, could have affected these results. Due to limitations of conducting work in a military environment, we were unable to record any information on potential participants that did not answer the survey; thus, we do not have information on non-response bias and cannot evaluate accordingly. Participants may or may not have been required to complete training on hazing in the military, which may affect these results.

Conclusions

Hazing presents in a myriad of ways and leads to negative outcomes for those affected and the organization. This research provides an understanding of this presentation by examining hazing experiences and outcomes through descriptive epidemiology in the U.S. military. Detailing this information can aid both leadership and practitioners in effectively selecting, tailoring, and implementing programming to address this issue. Further research should pilot potential solutions to this issue that affects both the individuals and the U.S. military.

Abbreviations

SMs

Service Members

US

United States

Author contributions

E.M. contributed to study design, completed analyses, and drafted the manuscript. D.W. contributed to analyses and manuscript drafting. A.B. contributed to data analysis. A.R. contributed to study design, manuscript review, and data collection. E.J.A. and L.M.O. contributed to study design, manuscript drafting, and analysis plans. C.B.C. oversaw data collection and study design, and contributed to the manuscript drafting. All authors reviewed the manuscript.

Funding

This project was funded by Congressionally Directed Research Programs. The funder did not have a role in the design, data collection, analysis, interpretation, or dissemination of these findings.

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to Department of Defense requirements but are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The study received approval by the Womack Human Research Protections Office and all participants provided electronic consent.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.