Military Sexual Assault, Post-Service Employment, and Transition Preparation among U.S. Military Veterans: New Directions for Research

ABSTRACT

Unlike studies of military sexual trauma (MST) among active-duty service members, most studies of veterans with MST have been clinical in nature, focused on estimating population prevalence rates, improving clinical responses or treatments, or have associated MST with subsequent health-related risk behaviors such as alcohol or drug abuse. The present study seeks to broaden our understanding of the corollaries of military sexual assault by considering the relationship between being a survivor of military sexual assault and post-service employment and transition experiences. Using secondary data from a survey of the members of a national organization for post-9/11 veterans, this exploratory study examines bivariate and multivariate relationships between self-reported experiences of military sexual assault and specific post-transition outcomes commonly used as indicators of a successful career transition from military to civilian life including employment status, time from transition to employment, and veterans’ perceptions of being prepared for their transition out of the military. Analysis from this study suggests that military sexual assault is impactful in the domains of employment and transition readiness. This paper’s findings point to a need to consider a range of inter-related social determinants of health and well-being when providing post-service career support for veterans.

KEYWORDS:

• Veterans
• military sexual assault
• military sexual trauma
• veteran employment
• military separation

Introduction

The problem of sexual assault and harassment within the military ranks is not a new one, having been studied in various forms since the 1960s. Institutional attention to addressing the problem of military sexual trauma (MST) increased significantly in the 1990s in part due to new policies, research, and efforts by Veterans Affairs (VA) to improve clinical screening and treatment (Allard et al., 2011; Purtle, 2016; Suris & Lind, 2008). Still, in the decades that followed, only disparate studies could be identified to establish a baseline prevalence for MST, with many studies lacking deliberate sampling or methodological consistency (Turchik & Wilson, 2010; Wilson, 2018). These limitations in defining the problem kept much of the focus of veteran-oriented research on estimating prevalence rates through clinical data and practice. From the beginning, much of this research has centered around psychological trauma, particularly among female soldiers and veterans (Himmelfarb et al., 2006; Martin et al., 2000) but also among men (Peterson et al., 2011), and has continued into recent years to largely focus on post-traumatic stress disorder (PTSD) or other health-related outcomes (Jaycox et al., 2022). Given the largely clinical focus of research into MST to-date, comparatively little research exists that examines the impacts of MST or military sexual assault on veterans outside of health-care contexts. The present study explores whether experiences of military sexual assault impact other areas of a veteran’s life, in this case, by examining employment and transition experiences based on a non-clinical, national sample of US veterans.

Research studies that specifically focus on veteran survivors of military sexual assault that go beyond direct trauma treatment are rare but are still largely medical in nature. Most recently, experiences of military sexual trauma have been linked to a wide variety of other adverse health and wellness outcomes such as addiction (Beckman et al., 2022), impact on sexual functioning in relationships (Blais & Livingston, 2021), or other risk behaviors (Forkus et al., 2021). One recent study did consider the impact of military sexual assault on the outcome of a vocational employment program, however this too was in the context of a broader medical rehabilitation program and thus not generalizable (Gross et al., 2022). Blais and Livingston (2021) utilize a non-clinical sample of service members or veteran women recruited via social media and identify psychological correlates linking MST to reduced sexual function, but this sample is limited in that it only examines partnered women. Due to the clinical (mostly VA-related) nature of most other data sources used to study MST, there are significant limitations on our ability to accurately measure and estimate the prevalence of MST among veterans in the general population (Wilson, 2018) and also to accurately understand the broader impacts of MST beyond for those in direct treatment. Wilson (2018) aptly points out that “more research is urgently needed using diverse methodologies … [because] our ability to provide adequate MST services to our military personnel and veterans hinges on accurate detection of the phenomenon” (593). While her focus is on improvements in treatment and primary prevention, many social services that support veterans across a range of needs should also consider how military sexual assault can be an important, yet not often considered, obstacle to successful outcomes.

The present study seeks to explore this gap in the research on how military sexual assault impacts outcomes for veterans beyond the better-studied contexts of physical and mental health. Extending the work of addressing sexual trauma beyond the clinical context is also a way to address the needs or experiences of the many “invisible” victims of sexual assault who do not come forward with their experiences, which data suggests may even be more common in the military (U.S. Commission on Civil Rights, 2013) than in other contexts where underreporting is also a serious concern. The prevalence of MST may also be underestimated in VA-related populations, given estimates that the majority of veterans are not enrolled in VA health care (Holder & Day, 2017), especially considering that individuals who experience sexual trauma may face disqualifying discharge or harbor mistrust toward military-related institutions. This paper’s analysis is exploratory in nature, conducting secondary analysis on a dataset from a national survey of mostly post-9/11 veterans about a variety of personal, political, and social issues. The intention of this paper is to explore the relationships between military sexual assault, post-service employment, and preparation for transition, providing direction for future research that integrates these domains into efforts to better understand and mitigate the impacts of military sexual trauma on veterans.

Background

Efforts to support veterans who had previously experienced military sexual assault as a service member include several important milestones, notably a 1992 mandate to the Department of Veterans’ Affairs (VA) naming military sexual trauma as an area of concern (Allard et al., 2011). More recently, a 2012 report commissioned by the Obama administration (U.S. Commission on Civil Rights, 2013) set the stage for a new round of reforms spanning the past decade. This internal effort was strengthened in response to high-profile incidents at military bases and VA health centers around the country and increasing public pressures to address sexual assault and harassment issues in the entertainment industry, professional workplaces, and college campuses during this same period. Most recently, the 2022 National Defense Authorization Act (NDAA) and a number of executive orders signed by President Biden have sought to specifically strengthen reforms within the Military Code of Justice and enact new procedural responses to sexual assault and harassment reports within the military (Mason, 2022). While these recent reforms are an important step forward for current service members, there are an estimated 18 million veterans including 3.7 million post-9/11 veterans (Vespa, 2020) who previously served during periods where much less support was offered to active duty survivors of military sexual trauma.

Early research examining sexual assault experiences among soldiers helped to set the agenda for understanding MST; namely, the focus was placed on the psychological effects of sexual trauma and the particularly negative impacts it had on female service members (Martin et al., 1998, 2000). Importantly, the experiences of sexual harassment in this early research were also linked to unit cohesion and combat effectiveness (Rosen & Martin, 1998), suggesting that sexual assault could have a direct impact on military effectiveness, leading to an increasing emphasis on sexual assault as a form of psychological trauma by U.S. legislators (Purtle, 2016). The implication of this work made MST an important consideration for other areas of concern, such as military readiness, retention and recruitment, both physical and psychological danger to service members (Szitanyi, 2020), and general well-being for active duty soldiers (Holland et al., 2016). This research and policy agenda linked military sexual assault to psychological trauma, moral injury, and physical combat risks, which has clear and urgent implications for many service-disabled veterans or those with mental health needs, however this narrow focus on military-specific forms of trauma (and the related research funding priorities supporting this work) led researchers to generally neglect inquiries into how other veterans might experience corollary impacts from sexual assault and harassment experienced during service.

A substantial part of the extant research on veterans focuses on the transition from military service to being a veteran. Theory on military transition has recognized the importance of addressing past trauma in order to develop a new self-narrative, building upon the idea that a meaningful work and career can be a mechanism for successful veteran transition (Haynie & Shepherd, 2011). Other theory considers the ecological context in which a veteran transitions as important for clinical interventions (Elnitsky et al., 2017). Within efforts to understand and support veterans’ transition, however, research studies that consider the role of military sexual assault are relatively sparse. These tends also focus only on women veterans, with very few studies considering socio-economic factors such as employment (Maya & Kimberley, 2018). Research has, however, begun to consider a wider range of social determinants of health, adding important context to a veteran’s transition needs and challenges. Notable in these efforts are research projects by The Veterans Metrics Initiative (TVMI) at Pennsylvania State University or the D’Aniello Institute for Veterans and Military Families (IVMF) at Syracuse University, whose studies highlight transition challenges such as navigating VA benefits or dealing with service-connected disabilities or examine benefits of a wide range of employment and career-preparation opportunities provided to veterans (Bradbard & Maury, 2021; Perkins et al., 2020). To date, however, almost no research exists which unites the clinically focused work on military sexual trauma with efforts to understand socio-demographic correlates of military separation and transition. The fact that the study of military sexual assault and its implications for veterans has so far been siloed into healthcare research represents a significant gap in the literature.

The few extant studies looking at military sexual assault in the context of employment or well-being outcomes, discussed below, clearly suggest that MST impacts outcomes. Research combining VA medical and employment program records has found that MST is linked to a lack of competitive employment at program completion (Gross et al., 2022). Moreover, this relationship may create a vicious cycle; research on adult trauma and PTSD consistently shows a relationship between low income or unemployment and dropping out of treatment programs (Matthieu & Ivanoff, 2006).

One key transition challenge for survivors of MST may be related to a heightened sense of institutional mistrust caused by the military’s response (or lack of response) to their assault experiences (Bourke, 2022). Once these service members become veterans, this mistrust may be extended to other governmental services such as the VA. Previous research has found that veterans who experienced MST, and particularly those who experienced retaliation for reporting, are less likely to access not only treatment but other needed support as well (Holland et al., 2016). Extant literature also suggests that sexual trauma can have other complex interactions with certain unique properties of military service. For instance, research among female service members has shown that many had experienced more instances of sexual trauma than other women, even prior to joining the military, itself an independent risk factor for subsequent adult sexual victimization (Martin et al., 1998; Turchik & Wilson, 2010). Outside of research specifically focused on veterans or service members, people with disabilities and particularly disabled women have been found to be more vulnerable to sexual victimization in national population-based studies (Basile et al., 2016; Mailhot Amborski et al., 2021). Therefore, it is not unexpected that there could also be connections between service-connected disabilities and subsequent risk of victimization among women veterans.

Despite the relative sparseness of research specifically linking military sexual assault or MST to veteran outcomes beyond clinical and health-care contexts, enough evidence exists in what research has been previously published to suggest that this area represents a substantial gap in our knowledge of how best to support veterans who are also survivors. The analyses in this paper focus on answering the question of whether there are differences between military sexual assault survivors and other veterans across variables measuring employment (both current and at the time of transition) and ratings of preparedness for their transition from the service that may provide direction for future research investigating how military sexual assault influences non-medical needs such as employment and transition support.

Methods

Sample and data collection

This study analyzed data collected by a recent national survey of predominately post-9/11 veterans, exploring several topics related to veterans’ service, post-transition experiences, and opinions on topics important to U.S. military veterans. A module on sexual assault and harassment was one of the topics surveyed. In the fall of 2021, Iraq and Afghanistan Veterans of America (IAVA) and the D’Aniello Institute for Veterans and Military Families at Syracuse University (IVMF) fielded the 2022 IAVA Membership Survey (Pritchard et al., 2022). The survey was open from September 2021 through October 2021, with a reminder e-mail each week (8 total) sent from representatives of IAVA. No incentives were provided to participants, and the survey was protected from unauthorized or bot participation by sending to a controlled distribution list rather than using an anonymous link. The study and the necessary confidentiality protocols were reviewed and approved as exempt by the Syracuse University IRB (IRB #21–245), and consent was obtained via the landing page of the online survey within Qualtrics. The online survey was sent to 111,634 valid e-mail addresses for members provided by IAVA, resulting in 7,417 started surveys (6.6%) and 5,524 completed surveys (74.5% completion rate). The final valid survey sample includes 5,157 participants, including 691 female veterans (14%), approximately proportional to the female population of recent US veterans (Vespa, 2020). For the purposes of this paper, the data presented will be based on the subsample of participants with a valid response (e.g., not left blank) on the military sexual assault screening item, who as a group do not significantly differ from the overall sample on basic demographics. The overall survey and analytic sample demographics are presented in Table 1.

Table 1. Survey and sample demographic comparison.

	Entire survey (N = 5,157)	Analysis sample (n = 4,407)
	n (%)	n (%)	p
Sex
Male	4203 (84.4)	3726 (84.5)	.682
Female	691 (13.9)	610 (13.8)	.991
Other	85 (1.7)	62 (1.4)	.247
Total valid	n = 4979	n = 4398
Age
18–30	54 (1.1)	43 (1.0)	.662
31–40	1497 (29.6)	1297 (29.4)	.887
41–50	1571 (31.1)	1357 (31.2)	.806
51–65	1497 (28.7)	1296 (29.4)	.868
Over 65	458 (8.9)	391 (8.9)	.769
Total valid	n = 5059	n = 4403
Race and ethnicity^a
White	3844 (77.6)	3445 (78.6)	.249
Black or African American	387 (7.8)	333 (7.6)	.696
Hispanic, Latino, or Spanish	569 (10.4)	486 (11.1)	.543
American Indian or Alaskan Native	199 (3.6)	167 (3.4)	.606
Asian	184 (3.4)	164 (3.7)	.946
Native Hawaiian or Pacific Islander	42 (0.8)	37 (0.8)	.984
Other	261 (4.8)	221 (5.0)	.620
Total valid	n = 4953	n = 4383
Veteran status
Veteran (yes)	4888 (94.8)	4168 (94.6)	.653
Post-9/11 veteran (yes)	4655 (90.3)	4002 (90.8)	.365
Total valid	n = 5157	n = 4407
Enlistment status
Enlisted	3734 (72.4)	3276 (76.2)	.192
Officer/warrant officer	1094 (21.2)	1024 (23.8)	.192
Total valid	n = 4828	n = 4300

^aThis item is coded as select all that apply, so counts indicate yes on a given category and the percentage reflects the percent of cases. Individuals endorsed a mean of 1.1 categories.

Measures

This paper uses items from throughout the survey to conceptualize the topics investigated in this analysis. Specifically, this survey makes use of the survey’s measures of military sexual assault and reporting experiences, measures of veteran employment outcomes, and items examining veterans’ perceptions of their level of preparedness for transitioning from military to civilian life.

Military sexual assault

The main screening item for military sexual assault was measured using the following binary (yes/no) question: “Are you a survivor of military sexual assault? (This includes things like unwanted touching, someone exposing themselves or unwanted sexual contact).” This measure is used throughout the analysis of specific topics to distinguish those veterans reporting prior military sexual assault experiences from those who did not; in other words, participants selected yes on this item if they self-identify as a survivor of military sexual assault due to their experiences. Table 2 provides results from this measure and follow-up questions, including veterans’ experiences with reporting sexual assault incidents and their perceptions of the military’s response.

Table 2. Military sexual assault survey items.

	Total	Male	Female	Sex diff.
	N (% total)	N (% males)	N (% females)	p
Experienced military sexual assault	526 (11.9)	151 (4.1)	358 (58.7)	.000*
If experienced (n = 526) did you report?	178 (33.8)	36 (23.8)	134 (37.4)	.002*
If not reported (n = 348) why not?
Did not think anything would be done	242 (69.9)	68 (60.2)	167 (74.6)	.007*
Fear of retaliation by my peers or chain of command	195 (56.4)	48 (42.5)	142 (63.4)	.000*
Concerned about impact on my career	189 (54.6)	52 (46.0)	132 (58.9)	.025*
Doubted that my chain of command would believe me	176 (50.9)	47 (41.6)	125 (55.8)	.014*
Concerned that my peers would treat me differently	171 (49.4)	65 (57.5)	102 (45.5)	.038*
Did not want to be involved in an investigation	161 (46.5)	47 (41.6)	111 (49.6)	.167
Worried I was less valued than perpetrator	117 (33.8)	27 (23.9)	86 (38.4)	.008*
Worried that people would think I was weak	114 (32.9)	47 (41.6)	66 (29.5)	.026*
Worried family/friends would treat me differently	57 (16.5)	29 (25.7)	26 (11.6)	.000*
Other	60 (17.3)	30 (26.5)	29 (12.9)	.002*
If reported (n = 178) did you experience retaliation?	112 (62.9)	17 (47.2)	90 (67.2)	.024*
View of trained military prosecutor rather than commander to handle complaint
More likely to report	267 (51.0)	60 (40.0)	196 (54.9)	.002*
Would not change my decision	242 (46.2)	82 (54.7)	156 (43.7)	.024*
Less likely to report	15 (2.9)	8 (5.3)	5 (1.4)	.011*

*p < .05.

For those indicating an experience of military sexual assault, a follow-up item asked, “Did you report the incident?,” which triggered additional branching logic. Those who did not report their assault experiences were given a multiple response list of prompts for “Why did you not report this?” (see Table 2). If they indicated that their assault was reported, survey participants were asked the yes/no follow-up question “Did you experience any kind of retaliation after reporting?” and also “If instead of your commander, a trained military prosecutor had the authority to make the decision to move forward with your case, which of the following impacts would it have had on your decision to report?” with response options of “I would have been more likely to report,” “I would have been less likely to report,” or “It would not have made a difference in my decision.”

Employment

Measures used for classifying veteran employment included three questions from the survey. Employment status was measured by asking “Which best describes your current employment situation?” with six response options ranging from “Employed Full Time (35+ hours per week)” to “Disabled or otherwise unable to work” (see Table 3). Employment at transition was measured with the question “How long were you without a job after transitioning out of the military?,” which included seven response options ranging from “Less than 3 months” to “Opted to attend school” (see Table 3).

Table 3. Key measures by military sexual assault survivor status.

	Military sexual assault
	Yes	No
	N (%)	N (%)	Sig. (p)
Current employment
Full-time	216 (44.2)	2091 (60.1)	.000*
Part-time	39 (8.0)	185 (5.3)	.009*
Unemployed, looking for work	33 (6.7)	170 (4.9)	.040*
Unemployed, not looking for work	25 (5.1)	81 (2.3)	.000*
Retired, not employed	48 (9.8)	509 (14.6)	.002*
Disabled or otherwise unable to work	128 (26.2)	443 (12.7)	.000*
Time until employment after transition
Less than 3 months	57 (15.2)	666 (28.8)	.000*
3–6 months	69 (18.4)	489 (21.1)	.119
Between 6 and 9 months	33 (8.8)	226 (9.8)	.285
Between 9 and 12 months	27 (7.2)	168 (7.3)	.490
Over 12 months	129 (34.5)	421 (18.2)	.000*
Never unemployed	8 (2.1)	79 (3.4)	.098
Opted to attend school	51 (13.6)	267 (11.5)	.121
How prepared were you to navigate
(1 = very unprepared, 5 = very prepared)
Transition from military to civilian life	2.52 (1.350)	3.00 (1.357)	.000*
Transition to post-service employment	2.41 (1.400)	3.00 (1.394)	.000*
Access to post-service education	2.92 (1.397)	3.24 (1.317)	.000*
Resources available in your local community	2.03 (1.253)	2.59 (1.298)	.000*
Healthcare and benefits system	2.47 (1.382)	2.90 (1.351)	.000*
Overall scale (Cronbach’s alpha = 0.878)	2.42 (1.023)	2.86 (1.098)	.000*
Mean length of service in years	6.87 (3.506)	7.18 (3.378)	.235

* p < .05.

Levene’s test indicates unequal variance in individual scale items, so p-values presented for those items are calculated using the more conservative, non-parametric Mann–Whitney U test.

Transition

Transition from military service to civilian life was measured using a scale. Five items asked about perceptions of preparation for specific aspects of transition (“Using the following scale, how prepared were you to successfully navigate the following …?”) on a 5-point Likert-type scale ranging from 1 = “very unprepared” to 5 = “very prepared” with 3 as neutral (Cronbach’s Alpha = 0.878). The transition topics measured are listed in Table 3.

In addition to these subjective measures of transition, the mean length of military service in years was calculated based on two other questions in the demographic section of the survey: “What year did you join the military?” and “What year did you separate from the military?”

Data analysis

The data analysis was conducted using IBM SPSS Statistics 27 and involved two types of analysis. First, t-tests comparing veterans who reported military sexual assault to veterans who did not were used to explore the bivariate relationships between study variables and military sexual assault. In certain circumstances (i.e., with ordinal items in the transition scale), additional tests such as Levene’s test for unequal variance and Mann–Whitney U are used and are indicated in the tables where applicable. In Table 2, t-tests are also used to compare military sexual assault experiences by sex. Second, regression analysis (multinomial logistic regression or ordinary least-squares regression as appropriate) is used to control for demographic variables while examining the relationships between military sexual assault experiences and employment or transition measures.

Results

Military sexual assault

Details of veterans’ experiences with military sexual assault are summarized in Table 2. Overall, 526 respondents or 12% of the sample self-identified as a survivor of military sexual assault, defined in the question as “having experienced sexual assault or things like unwanted touching, someone exposing themselves, or unwanted sexual contact.”

Most participants self-identifying as a survivor of military sexual assault are female (70%) and survivors comprise 4% of male survey participants and 59% of female participants. Among male survivors, 24% reported the incident, and about half of those who reported (47%) said they experienced some form of retaliation after reporting. Among female survivors, 37% reported the incident and 67% said they experienced retaliation for having reported. This high rate of retaliation among those who did report their assault is commensurate with the perceptions of the 66% of survivors overall who did not report the incident, 56% of whom did not do so out of “fear of retaliation by my peers or chain of command.” Table 2 also shows veterans’ opinions about having an independent investigator outside of their chain of command handle complaints. After this survey was conducted, a version of this proposed change was implemented in the 2022 National Defense Authorization Act (NDAA) and will go into effect December 27, 2023, as Public Law 117–81.

Employment

Data examining the relationship between experiences of military sexual assault and current employment, as well as the time until securing employment upon transitioning out of the military, are presented in Table 3. Current employment status significantly differed between military sexual assault survivors and other veterans. Military sexual assault survivors are less often employed full time (44.2% vs 60.1%, p = .000), less often retired (9.8% vs 14.6%, p = .002), and are disabled and not working almost twice as often as other veterans (26.2% vs 12.7%, p = .000).

To determine if these relationships are significant when controlling for demographic differences, current employment status was further analyzed using multinomial logistic regression. Using full-time employment (>36 hours) as the reference category, military sexual assault was considered in a model using dummy variables to control for sex (1 = male), age (1 = over 40 years old), and officer/enlisted status (1 = officer). Race was measured using a multiple-response item, included in models by coding as 1 if the racial/ethnic category was endorsed and 0 if that category was not endorsed. For example, individuals who endorsed both white and Hispanic are coded as both white = 1 and Hispanic = 1 although most individuals endorsed only one category (mean is 1.1 endorsed categories), and only individuals endorsing at least one category are included in the analysis. For brevity, only the results from this model for categories where military sexual assault is significant are presented in Table 4. When these controls are considered, military sexual assault remains a significant predictor for both unemployed categories and for disability, but is no longer significant for veterans who are not working due to retirement. Relative to veterans who are full-time employed, unemployed veterans (looking for work) are 2.2x more likely to be military sexual assault survivors (exp(B) = 2.190), unemployed veterans (not looking for work) are 2.2x more likely to be military sexual assault survivors (exp(B) = 2.186), and veterans who are disabled or otherwise unable to work are 3.7x more likely to be survivors of military sexual assault (exp(B) = 3.668).

Table 4. Multinomial logistic regression model comparing current employment status and military sexual assault (reference category = full-time employment).

	B	S.E.	Sig.	EXP(B)	CI – low	CI – High
Unemployed, looking for work
Intercept	−2.321	0.372	0.000
Sex (male = 1)	0.334	0.275	0.225	1.396	0.814	2.394
Age (over 40 years old = 1)	−0.002	0.163	0.990	0.998	0.725	1.373
Enlistment (officer/warrant = 1)	−0.170	0.204	0.405	0.844	0.566	1.258
White	−0.701	0.244	0.004	0.496	0.307	0.800
Black	0.427	0.295	0.148	1.532	0.859	2.732
Hispanic	−0.011	0.270	0.966	0.989	0.583	1.677
Asian, Hawaiian, or Pac Islander	−0.296	0.388	0.447	0.744	0.348	1.593
Amer Indian or Alaskan	0.074	0.409	0.856	1.077	0.483	2.402
Military sexual assault survivor	0.784	0.270	0.004	2.190	1.290	3.717
Unemployed, not looking for work
Intercept	−2.867	0.474	0.000
Sex (male = 1)	−0.486	0.321	0.130	0.615	0.328	1.154
Age (over 40 years old = 1)	0.086	0.218	0.693	1.090	0.711	1.669
Enlistment (officer/warrant = 1)	−0.419	0.286	0.144	0.658	0.375	1.153
White	0.016	0.344	0.963	1.016	0.518	1.995
Black	0.225	0.447	0.616	1.252	0.521	3.009
Hispanic	0.278	0.360	0.440	1.320	0.652	2.675
Asian, Hawaiian, or Pac Islander	0.263	0.474	0.579	1.301	0.514	3.292
Amer Indian or Alaskan	−0.074	0.604	0.902	0.928	0.284	3.033
Military sexual assault survivor	0.782	0.334	0.019	2.186	1.136	4.204
Disabled or otherwise unable to work
Intercept	−2.147	0.250	0.000
Sex (male = 1)	0.380	0.179	0.034	1.463	1.030	2.078
Age (over 40 years old = 1)	0.744	0.114	0.000	2.105	1.683	2.631
Enlistment (officer/warrant = 1)	−0.870	0.147	0.000	0.419	0.314	0.559
White	−0.265	0.163	0.104	0.767	0.557	1.056
Black	0.296	0.207	0.153	1.345	0.896	2.018
Hispanic	−0.069	0.181	0.704	0.934	0.655	1.331
Asian, Hawaiian or Pac Islander	−0.347	0.279	0.214	0.707	0.409	1.221
Amer Indian or Alaskan	0.891	0.210	0.000	2.439	1.617	3.677
Military sexual assault survivor	1.300	0.171	0.000	3.668	2.624	5.128

Nagelkerke’s R-squared = 0.152. Significance of p < .05 indicated in bold. Full model also considers part-time employment and retired or no longer working. For brevity, these results are not presented as military sexual assault survivor is non-significant for these categories.

Table 3 also reports on the time elapsed between separation from the military and securing employment reported by veterans. Overall, 24.1% of those veterans experiencing military sexual assault had a job secured before leaving the military, compared to 33.5% of those who had not experienced military sexual assault (p = .000). Veterans without military sexual assault experiences had secured employment in less than 3 months after transition at nearly twice the proportion of those who experienced assault (28.8% vs 15.2%, p = .000), whereas the opposite is true of those taking over 12 months after transition to find employment. More than one-third (34.5%) of survivors of military sexual assault reported going more than 12 months before securing employment after transition compared to only 18.4% of other veterans (p = .000).

Once again, follow-up analysis using multinomial logistic regression was used to control for demographic variables to better validate these apparent differences. This analysis compared individuals who reported some delay to employment (categories “Less than 3 months” through “Over 12 months”) and excluded those with no employment gap or those who attended school. Ordinal regression was ruled out for this analysis due to uneven distribution of cumulative probabilities for the categories or violations of parallel line tests when appropriate link functions were considered. In such cases, multinomial regression is advised as an alternative (Hosmer et al., 2013). Once again for brevity, Table 5 presents only results from this model for the category “Over 12 months” where military sexual assault was a significant category predictor. Consistent with the bivariate results even when controlling for available demographics, veterans who took over 12 months to secure employment were 2.6x more likely to be military sexual assault survivors than those securing employment in less than 3 months (exp(B) = 2.566).

Table 5. Multinomial logistic regression model comparing the time from military transition to securing employment and military sexual assault (reference category = less than 3 months).

Over 12 months	B	S.E.	Sig.	EXP(B)	CI – low	CI – High
Intercept	0.195	0.299	0.514
Sex (male = 1)	−0.399	0.210	0.057	0.671	0.445	1.012
Age (over 40 years old = 1)	−0.413	0.130	0.002	0.662	0.513	0.854
Enlistment (officer/warrant = 1)	0.476	0.160	0.003	1.610	1.176	2.203
White	−0.270	0.203	0.183	0.763	0.513	1.136
Black	0.607	0.281	0.031	1.836	1.057	3.187
Hispanic	0.209	0.223	0.348	1.233	0.796	1.908
Asian, Hawaiian, or Pac Islander	0.412	0.346	0.234	1.511	0.766	2.978
Amer Indian or Alaskan	0.556	0.284	0.051	1.744	0.999	3.044
Military sexual assault survivor	0.942	0.219	0.000	2.566	1.670	3.943

Nagelkerke’s R-squared = 0.066. Significance of p < .05 indicated in bold. Full model also considers 3–6 months, between 6 and 9 months, and between 9 and 12 months. For brevity, these results are not presented as military sexual assault survivor is non-significant for these categories.

Transition

Veterans’ ratings of preparedness for transition on a scale examining elements such as employment, education, community resources, and the healthcare system are also presented in detail in Table 3. Similar to bivariate comparisons on employment measures that showed significantly worse employment outcomes for military sexual assault survivors, all transition preparation items were rated significantly higher by veterans who indicated no military sexual assault experiences. Service members experiencing military sexual assault had, on average, slightly shorter service length than other veterans (6.87 vs. 7.18 years), but this difference was not statistically significant (p = .235). Time in service also had relatively large standard deviations (3.506 and 3.378, respectively), reflecting a wide range of service experiences in general that may defy a simplistic explanation.

Based on the extant literature, this study considers the possibility that both military sexual assault and time-in-service might impact transition preparation (e.g., hypothesizing that longer service is more likely to be planned, and thus better prepared where shorter service may result from unplanned or early departure). As such, both years in service and military sexual assault were considered possible predictors of transition preparedness. Once again, a regression model was employed as a follow-up analysis to further validate these apparent differences using appropriate demographic controls for sex, race, age (over/under 40 years old), and officer/enlisted status. This time, ordinary least squared regression was used due to the nature of the dependent variable (i.e., a Likert-type scale). Models are presented first for control variables only, next adding years of service, then third adding military sexual assault (Table 6). Both independent variables are statistically significant when added, and years of service are statistically significant in both models. Years of service are positively associated with better preparation (0.04 increase in the scale mean per service year), while military sexual assault is negatively associated with better preparation for transition (0.44 decrease in the scale mean if a survivor). Also notable is that being an officer (rather than enlisted) was a strong positive predictor of transition preparation (0.39 increase in the scale mean if an officer), while American Indian/Alaskan Native ethnicity was a strong negative predictor of transition preparedness (0.44 decrease in the scale mean if American Indian/Alaskan Native).

Table 6. OLS regression model transition preparedness scale versus years of service and military sexual assault experiences.

	MODEL 0 – Controls			Model 1 – years of service			Model 2 – military sexual assault
	B	t	Sig.	B	t	Sig.	B	t	Sig.
Constant	2.311	17.150	0.000	2.076	13.950	0.000	2.350	14.374	0.000
Male	0.102	1.161	0.246	0.084	0.958	0.338	−0.170	−1.570	0.117
Over 40 years old	0.033	0.428	0.669	0.004	0.057	0.955	−0.003	−0.037	0.970
Officer/warrant officer	0.441	3.544	0.000	0.396	3.188	0.001	0.393	3.186	0.001
White	0.159	1.427	0.154	0.164	1.481	0.139	0.163	1.488	0.137
Black	0.234	1.508	0.132	0.238	1.543	0.123	0.239	1.560	0.119
Hispanic	0.011	0.098	0.922	0.017	0.150	0.881	0.008	0.066	0.947
Asian, Hawaiian or Pac Islander	0.037	0.249	0.803	0.029	0.195	0.846	−0.018	−0.121	0.904
Amer Indian or Alaskan	−0.427	−2.304	0.021	−0.441	−2.391	0.017	−0.443	−2.419	0.016
Years in service				0.037	3.631	0.000	0.038	3.799	0.000
Military sexual assault survivor							−0.442	−3.920	0.000
F	3.112			4.263			5.426
R-squared	0.023			0.035			0.049
Adj. R-squared	0.016			0.027			0.040
R-squared change	0.023			0.012			0.014

N = 1,064. Significance of p < .05 indicated by bold, and p < .01 indicated by bold and italics.

Discussion

The data presented in this paper explore post-service correlates of military sexual assault based on an exploratory analysis of data from a national sample of U.S. military veterans who participated in a wide-ranging survey through one of the nation’s largest membership organizations for post-9/11 veterans. Unlike most other research on veterans and military sexual assault, which focuses on clinical encounters with military sexual trauma (MST), this study seeks to highlight important associations beyond those typically examined through clinical or a service-seeking population. It is, therefore, the goal of this paper to highlight areas in which military sexual assault seems to have a differential impact on a veteran’s post-service experiences, and direct future research toward deeper investigations into the mechanisms behind, and ultimately opportunities to respond to, these apparent relationships.

Overall, this study’s findings are aligned with previous research indicating that military sexual assault is common, particularly among women; around 4.1% of male and 58.7% of female veterans indicated having such an experience. Prevalence rates are high in this sample, around twice that of studies with assault-only measures that exclude harassment (1.9% of men and 23.6% of women) and closer to measures that do include sexual harassment as well as assault (8.9% of men and 52.5% of women) as calculated via meta-analysis (Wilson, 2018). One possibility is that participants in the present study responded based on any adverse sexual experiences in the military and includes many types of harassment due to the nonspecificity of the measure (“Are you a survivor”) used in the survey. Another possibility is that this sample differs from most other MST research (typically VA-served clinical populations) because it samples from (1) a non-clinical population that is (2) mostly post-9/11 veterans. Notably, official estimates of the proportion of veterans served by VA health care are as low as around one-third (Holder & Day, 2017), meaning that most veterans are not captured in VA clinical studies. While this study is not designed to estimate prevalence, nevertheless the high prevalence rate found here supports Wilson’s (2018) urgent call to develop better population-based studies to estimate the scope of military sexual assault within veteran and active duty populations. Extant research also raises the possibility that early or bad discharge status can be related to assault experiences and can disqualify veterans from benefits like VA care and that institutional mistrust fostered by poor institutional response may contribute to a veteran’s unwillingness to engage VA health care, and therefore, survivors may not be well represented in clinical VA samples (Bourke, 2022; Brignone, 2017; Holland et al., 2016). Considering other well-documented barriers to support for those experiencing sexual assault within the military, the data presented in Table 2 reflect the reality that most of those experiencing sexual assault in the military have not, by and large, felt well supported at the time of the incident. Experiences of sexual victimization, subsequent complex trauma, and associated lack of institutional support contribute to ongoing disadvantages for many veterans to whom these things happened during their military service (Bourke, 2022). Given the high rates of reported retaliation and the historical lack of institutional responsiveness, being a “squeaky wheel” within the military may carry additional unmeasured consequences including bad discharge status, lack of professional references and separation support, or other stressful stigma that may increase transition difficulties (Brignone, 2017; Holland et al., 2016). Especially for women, research has found that unwanted sexual experiences are an explanatory factor in negative impacts on service members’ health (Jaycox et al., 2022; Schram et al., 2022). These impacts likely extend far beyond physical or mental health and may have different implications by gender; however, to date, little research has been conducted that demonstrates possible impacts outside of clinical populations. This study’s exploratory analysis suggests that a military sexual assault can have deleterious impacts for veterans in non-clinical domains, in this case, employment and transition preparedness, as they separate from the service and possibly beyond.

Data from this study show that veterans who experienced military sexual assault during their time in the service had considerably higher barriers to obtaining employment than did other veterans. This interpretation is further supported by their subjective ratings of the transition experience, most notably perhaps in terms of feeling less prepared for life after separation. Military sexual assault experiences and shorter time-in-service are possible mechanisms that may explain this lack of preparation, although the low explanatory power of these models (and the extant literature) suggest that there are almost certainly additional factors not considered here which may also have an impact on transition readiness. While the present analysis investigates employment impacts at transition, future research should consider how enduring these employment effects may be over time. The fact that military sexual assault was also associated with lower rates of retirement and higher rates of unemployment due to disability offers support to previous research identifying connections between military sexual assault and combat-related injury (Szitanyi, 2020), and suggests that these connections are deserving of further research inquiry.

It is possible that in addition to clinically identified cases where military sexual assault contributes to MST, unreported experiences of military sexual assault or harassment may nevertheless impact a veteran’s ease-of-transition and employment success via the mechanisms of an abridged military career or disrupted career transition processes. This implication is difficult to parse out since separation can be voluntary, involuntary, or through disability that itself may or may not be more likely because of, or even directly related to, military sexual assault experiences (Brignone, 2017). Data from this survey are enough, however, to recommend this hypothesis be further explored given that exploratory multivariate analyses with demographic controls showed a significant influence on transition preparedness for both time-in-service and military sexual assault. Any such inquiry would likely require a carefully designed qualitative research study since there is unlikely to be accessible, honest, quantitative data with enough detail to explore the nuanced connection between stigmatizing experiences and the personal reasons service members choose to separate.

Another theoretical possibility that may explain the ongoing impact of military sexual assault on the lives of some veterans, including employment and transition difficulties, may also lie in the linkage between sexual assault and subsequent service-connected disability (Suris & Lind, 2008; Szitanyi, 2020). Sequelae of trauma such as PTSD or anxiety occurring after sexual assault experiences may contribute to a veteran’s disability status, either directly (e.g., as a mental health diagnosis), indirectly through physical injuries or mistakes made while distracted by complex trauma occurring within a high-risk work environment (e.g., deployment and training), or both (Allard et al., 2011). Given the high-stress environments in which military service members regularly operate and findings that a large proportion of military sexual assault survivors do not seek help due to actual retaliation or a perception that they will experience retaliation, there is a strong possibility that military sexual assault also increases the risk of seemingly unrelated incidents or injuries. It is possible that military sexual assault is both a cause and risk factor for some significant types of service-related injuries, trauma, or disabilities.

Limitations

The data used in this study are based on a national survey of members of an organization advocating for the needs of post-9/11 veterans. Typical limitations of self-selection in this type of survey research are present in this data (e.g., voluntary participation, overrepresentation of post-9/11 veterans, possible bias toward issues advocated by the organization), including the fact that many of the measures are less detailed than they would be in a purposive academic study owing to the reason for this survey and wide range of topics covered. This survey was conducted during the fall of 2021. Therefore, the impact of the COVID-19 pandemic cannot be specifically estimated on certain variables (e.g., current employment) or for any very recently separated veterans whose subjective experience of transition may be shaped by having occurred during the pandemic (3.1% of veterans surveyed separated in 2020 or 2021). All things considered these disruptions are likely to be minimal since one would not expect the pandemic to impact military sexual assault survivors in a different way than all other veterans.

This study relies on a single, broad question to measure a service member’s experience of military sexual assault. This type of nonspecific question does not classify the types or severity of experiences that other behaviorally specific victimization measures would provide, but as-is this item prioritizes a survey participant’s self-identification as a survivor regardless of their specific experiences. Since this study seeks to explore enduring impacts of military sexual assault experiences on post-service employment and transition, a measure that captures a participant’s self-identification with having had sexual assault experiences still arguably measures experiences impactful enough for participants to self-identify as a survivor, and therefore it is reasonable to use this measure for the purposes indicated.

Conclusion

The exploratory analysis presented in this paper has been undertaken to inform and direct future research toward investigating the impacts of military sexual assault on veterans’ post-service lives beyond just clinical concerns. In contrast and as a supplement to the body of work linking veterans’ history of MST to physical and mental health outcomes, this paper highlights the need to address the post-service consequences of military sexual assault across other domains such as veteran employment and transition readiness. Military sexual assault experiences can have deleterious impacts on areas outside of health and can do so even if these experiences are not reported or disclosed to health care providers. This paper utilized data from a national survey of post-9/11 veterans, placing military sexual assault alongside other key issues facing veterans and was thereby able to identify and empirically demonstrate associations between this and other important parts of veterans’ transition and post-service lives. Although this paper explored topics of post-service employment and transition, there are undoubtedly other areas or aspects veterans’ experiences (e.g., higher education or disability) where additional research could produce unique insights, including but not limited to replication among other groups of veterans (i.e., pre-9/11, non-US veterans). A comprehensive approach to understanding how specific military experiences and veteran well-being are related should be standard practice in all research intended to support veterans’ success. Most services for veterans already recognize the impacts of highly visible military experiences (e.g., service-connected injuries, PTSD, and family separation) on veterans’ needs, but experiences such as military sexual trauma or intimate partner violence that are more often “hidden” (for personal, cultural, or structural reasons) are just as important barriers to future well-being and are deserving of similar consideration when designing interventions. The data presented in this paper represent a first step in that direction, demonstrating to researchers and practitioners the possibilities of a more integrated research agenda in support of all veterans.

Acknowledgments

The author would like to thank Walter DeKeseredy and Deborah Bradbard for their feedback on earlier versions of this manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).