National Prevalence of Sexual Violence by a Workplace-Related Perpetrator

Abstract

Introduction

Workplace sexual violence is not a new phenomenon but has received increased attention recently with the re-emergence of the #metoo movement. Gaps exist in the understanding of the prevalence of this problem in the U.S., its perpetrators, and its impacts.

Methods

Using 2010−2012 data from the National Intimate Partner and Sexual Violence Survey (22,590 women and 18,584 men), this study examined the prevalence of several types of sexual violence by a workplace-related perpetrator (authority figure or nonauthority figure) and numerous impacts of the violence, including psychological impacts, safety concerns, and missing days of work or school. Data were analyzed in 2018.

Results

In the U.S., 5.6% of women (almost 7 million) and 2.5% of men (nearly 3 million) reported some type of sexual violence by a workplace-related perpetrator. Almost 4% of women (3.9%) reported sexual violence by nonauthority figures and 2.1% reported authority figures; 2.0% of men reported sexual violence by nonauthority figures, and 0.6% reported authority figures. For women, the most commonly reported sexual violence type was unwanted sexual contact (3.5% of women); for men, it was noncontact unwanted sexual experiences (1.3% of men). An estimated 1 million women (0.8%) have been raped by a workplace-related perpetrator. For women and men, fear was the most commonly reported impact of workplace-related sexual violence.

Conclusions

These findings suggest that workplace prevention efforts that do not address different components of workplace harassment may not be adequate to address all forms of sexual violence occurring across the U.S. in the workplace context.

INTRODUCTION

Recent public allegations of sexual misconduct in the workplace started in Hollywood¹ but quickly reverberated across industries.² In fall 2017, the #metoo movement³ re-emerged out of high profile celebrity cases in the news, becoming a national conversation and viral hashtag. Bringing sexual violence (SV) and its impacts to the forefront has helped destigmatize victimization and encourage systemic change (metoomvmt.org), particularly in the workplace.⁴ The #metoo movement has also generated recent attention in the research community for a renewed focus on workplace sexual harassment as a public health issue.⁵ However, SV in the workplace is not novel; it has plagued workers for centuries. It was not until the widely cited book, Sexual Harassment of Working Women,⁶ was published in 1979 that workplace sexual harassment became recognized in the courts as a form of sex discrimination. Much has been learned, but major gaps remain, such as the U.S. prevalence, perpetrators, and impacts of workplace SV.

The term “SV” is defined as unwanted penetration by use of force or alcohol/drug facilitation (e.g., rape), pressured sex (e.g., sexual coercion), unwanted sexual contact (e.g., groping), and noncontact unwanted experiences (e.g., unwanted sexual remarks) and may be considered “sexual harassment” in certain settings, such as workplaces.⁷ Numerous terms are used in the literature to describe workplace SV, such as workplace sexual harassment,⁸⁻¹⁰ workplace violence,¹¹ and workplace sexual abuse.¹² In this paper, the term workplace-related SV (WRSV) is used to encompass any unwanted sexual experiences. The most frequently used legal definition of sexual harassment is from the Equal Employment Opportunity Commission: “unwelcome sexual advances, requests for sexual favors, and other verbal or physical harassment of a sexual nature” that is so frequent or severe that it creates a hostile or offensive work environment or results in an adverse employment decision (i.e., quid pro quo).^13,14 Using behaviorally specific questions to measure sexual harassment/assault, a recent online panel survey found that 38% of women and 13% of men experienced workplace sexual harassment in their lifetimes.¹²

Research indicates that exposure to WRSV occurs as early as adolescence.^11,15,16 Despite women comprising majority of the sexual harassment claims,^17,18 men are also victims. From fiscal year 2010 to 2018, the most recent year of data collected by the Equal Employment Opportunity Commission (excluding Fair Employment Practices Agencies claims), the total number of sexual harassment claims declined by 4.2%, but a consistent number of claims had been filed by men over this period (16.2% in fiscal year 2010 and 15.9% in fiscal year 2018).¹⁹

Job satisfaction, organizational commitment, and health (e.g., serious injury, anxiety) are impacted adversely by WRSV.^10,20 It has also been associated with costs to the organization, including legal fees, high turnover, reduced productivity, negative publicity, and increased absenteeism.^{8–10,21–25}

The type of SV acts perpetrated in the workplace are also important considerations, particularly as they relate to the impact on a victim. Research suggests that SV behaviors involving physical contact that occur less frequently but are perceived to be more severe (e.g., coercion to have sex) have similar negative effects on occupational well-being as nonphysical but more frequent harmful workplace experiences (e.g., gender harassment).^26,27

An important aspect of preventing WRSV is understanding the perpetrators. Previous literature has found that, for female victims, workplace-related perpetrators are usually male, but the literature is mixed regarding male victims.^28–30 In addition, there may be different impacts of WRSV that are perpetrated by an authority figure versus a coworker, given the inherent power differential in a boss−employee relationship. The recent national study found that women (25%) were more likely than men (10%) to report a boss/supervisor as the perpetrator of sexual harassment/assault, whereas approximately 30% of both sexes reported a coworker.¹²

This study fills gaps in the literature using a large, nationally representative telephone survey sample of U.S. adults (>41,000 respondents) to examine 5 forms of SV by a workplace-related perpetrator. The prevalence for women and men of each form of SV by workplace-related perpetrators is reported, including 2 types: (1) a supervisor, boss, or superior in command (authority figure) or (2) a coworker, customer, or client (nonauthority figure), which provides more detail on the types of perpetrators than previous studies. The proportion of female and male victims by the sex of perpetrator and proportion of victims who experienced impacts resulting from WRSV is described. Stratifying the data and reporting the results in this manner will increase understanding of the sex of perpetrators of male victims, as well as unique impacts associated with WRSV by authority and nonauthority figures.

METHODS

Study Sample

Data were from the Centers for Disease Control and Prevention’s 2010−2012 National Intimate Partner and Sexual Violence Survey (NISVS). NISVS is an ongoing, nationally representative, random-digit-dial telephone survey of U.S. noninstitutionalized individuals aged ≥18 years using a dual-frame sampling strategy (landline and cell phone). The survey is conducted in all 50 states and the District of Columbia, providing information on violence victimization of adult women and men. A total of 22,590 women and 18,584 men who completed the survey in 2010−2012 comprised the study population. Using the American Association for Public Opinion Research response rate 4,³¹ the response rate ranged from 27.5% to 33.6%, and the cooperation rate ranged from 80.3% to 83.5%, indicating a high percentage of participation among adults who were contacted and determined to be eligible. Weights that account for sampling procedure, nonresponse, and phone coverage are available to generate nationally representative estimates for the U.S. adult population. More information is provided in Smith et al.³²

Measures

Based on behaviorally specific questions, multiple forms of SV victimization were measured, including rape (completed forced penetration, attempted forced penetration, and alcohol/drug-facilitated completed penetration), being made to sexually penetrate someone (completed, attempted, and alcohol/drug-facilitated), sexual coercion (i.e., nonphysically pressured unwanted penetration), unwanted sexual contact (e.g., groping), and noncontact unwanted sexual experiences (e.g., exposed sexual body parts) (Appendix Table 1, available online). Experience of SV victimization over the participant’s lifetime was examined. Given the low disclosure of being made to penetrate by women and rape victimization by men, rape of women and being made to penetrate of men is reported.

This analysis only focused on workplace-related perpetrators. Individuals who experienced one or more of the types of SV were asked the type of perpetrator at the time when the violence by the perpetrator happened the first time. If the perpetrator was a workplace authority figure (including boss, supervisor, or superior in command) or nonauthority figure (including coworker, customer, or client), the type of perpetrator was coded as workplace related. Some victims may have experienced SV by both types of workplace-related perpetrators.

Respondents who reported any victimization (not limited to SV and could also be stalking or physical violence by an intimate partner) by a specific perpetrator were asked about the impact of any of the violence committed by that perpetrator. For this study, the following impacts were examined among respondents who experienced SV by a workplace-related perpetrator: missing at least one day of work/school, fear, concern for one’s own safety, post-traumatic stress disorder (PTSD) symptoms (e.g., nightmares), and injury (Appendix Table 2, available online).

Statistical Analysis

Data were analyzed in 2018. The outcome measures included: (1) lifetime prevalence of SV overall and by subtype by a workplace-related perpetrator among U.S. women and men, (2) proportion of male and female perpetrators among victims of WRSV, and (3) proportion of female and male victims of WRSV who reported certain impacts. Outcome measures were stratified by victim sex and workplace perpetrator (nonauthority figure or authority figure). The analyses were conducted using SUDAAN, version 11.0.1 to account for the complex sampling design and to incorporate weights to derive nationally representative estimates.

RESULTS

In the U.S., 5.6% of women (an estimated 7 million women) reported some type of SV by a workplace-related perpetrator in their lifetime; 3.9% reported SV by a nonauthority figure, and 2.1% reported SV by an authority figure (Table 1). For women, the most commonly reported type of lifetime SV by a workplace-related perpetrator was unwanted sexual contact, reported by 3.5% of U.S. women (2.3% of women reported nonauthority figure, and 1.4% of women reported authority figure-perpetrated unwanted sexual contact). Noncontact unwanted sexual experiences perpetrated by a workplace-related perpetrator were reported by 2.4% of U.S. women (1.8% by a nonauthority figure and 0.7% by an authority figure). Approximately 0.8% or 1 million women are estimated to have been raped and a similar number sexually coerced by a workplace-related perpetrator in their lifetimes. The rape of women by a workplace-related perpetrator appears to be more commonly perpetrated by a nonauthority figure (Table 1).

Table 1.

Lifetime Sexual Violence^a of U.S. Women and Men by Workplace-Related Perpetrator,^a NISVS

Type of violence	Any workplace perpetratora		Nonauthority figure (Coworker, customer, client)a		Authority figure (Supervisor, boss, superior in command)a
	Estimated number of victims	Weighted % (95% CI)	Estimated number of victims	Weighted % (95% CI)	Estimated number of victims	Weighted % (95% CI)
U.S. women
Any sexual violencea	6,775,000	5.6 (5.2, 6.1)	4,709,000	3.9 (3.5, 4.3)	2,496,000	2.1 (1.8, 2.3)
Rape	940,000	0.8 (0.6, 0.9)	747,000	0.6 (0.5, 0.8)	201,000 0.2	0.2 (0.1, 0.2)
Sexual coercion	909,000	0.8 (0.6, 0.9)	413,000	0.3 (0.2, 0.5)	505,000	0.4 (0.3, 0.5)
Unwanted sexual contact	4,169,000	3.5 (3.1, 3.8)	2,749,000	2.3 (2.0, 2.6)	1,635,000	1.4 (1.1, 1.6)
Noncontact unwanted sexual experiences	2,894,000	2.4 (2.1, 2.7)	2,135,000	1.8 (1.5, 2.0)	859,000	0.7 (0.5, 0.9)
U.S. women
Any sexual violencea	2,868,000	2.5 (2.2, 2.8)	2,300,000	2.0 (1.7, 2.3)	678,000	0.6 (0.4, 0.8)
Made to penetrate	184,000	0.2 (0.1, 0.2)	157,000	0.1 (0.1, 0.2)	-b	-b
Sexual coercion	414,000	0.4 (0.2, 0.5)	220,000	0.2 (0.1, 0.3)	196,000	0.2 (0.1, 0.3)
Unwanted sexual contact	1,398,000	1.2 (1.0, 1.5)	1,064,000	0.9 (0.7, 1.1)	354,000	0.3 (0.2, 0.4)
Noncontact unwanted sexual experiences	1,501,000	1.3 (1.1, 1.5)	1,249,000	1.1 (0.9, 1.3)	315,000	0.3 (0.2, 0.4)

^aAvictim may have experienced more than one type of sexual violence or a given type of sexual violence by more than one type of perpetrator.

^bUnstable estimates are suppressed: relative SE >30% or cell size ≤20.

NISVS, National Intimate Partner and Sexual Violence Survey.

Approximately 2.5% of U.S. men (an estimated 3 million men) reported lifetime SV by a workplace-related perpetrator; 2.0% reported SV by a nonauthority figure, and 0.6% reported SV by an authority figure (Table 1). Unlike women, the most common type of lifetime SV against men by a workplace-related perpetrator was noncontact unwanted sexual experiences, reported by 1.3% of U.S. men (1.1% by a nonauthority figure and 0.3% by an authority figure). Unwanted sexual contact was reported by 1.2% of U.S. men (0.9% by a nonauthority figure and 0.3% by an authority figure). More than an estimated 400,000 men (0.4%) have been sexually coerced in their lifetime by a workplace-relate perpetrator, and an estimated 184,000 (0.2%) have been made to penetrate (Table 1).

Most of the female victims of WRSV reported only male perpetrators (96.2%). For male victims of WRSV, the findings were split between only male (40.7%) and only female perpetrators (53.6%). This pattern was observed for male victims with authority figure and nonauthority figure perpetrators. Very few female or male victims reported both male and female perpetrators (Table 2).

Table 2.

Sex of Workplace-Related Sexual Violence Perpetrators^a Among Victims, NISVS

Sex of perpetrator	Any workplace SV perpetrator	Nonauthority figure (Coworker, customer, client)	Authority figure (Supervisor, boss, superior in command)
Among female victims
Male only (%)	96.2	95.4	97.8
Female only (%)	2.8		–b
Both male and female (%)	–b	–b	–b
Among male victims
Male only (%)	40.7	36.9	56.8
Female only (%)	53.6		41.4
Both male and female (%)	5.5	5.8	–b

Note: Based on unweighted perpetrator data.

^aA victim may report multiple perpetrators.

Unstable estimates are suppressed: relative SE >30% or cell size ≤20.

^bNISVS, National Intimate Partner and Sexual Violence Survey; SV, sexual violence.

The most commonly reported impact among female SV victims by a workplace-related perpetrator was fear: almost 30% of female victims (29.8% or an estimated 2 million victims) reported fear, similarly for victims perpetrated by an authority figure (30.9%) or nonauthority figure (28.3%) (Table 3). Concern for safety and PTSD symptoms were also commonly reported by female victims of WRSV—reports of these impacts by female victims ranged from approximately 19% to 23% because of violence by authority figures or nonauthority figures from the workplace. Missing work/school and injury were less commonly reported, but approximately 7% of victims (almost 0.5 million victims) reported missing work/school, and 3.6% reported injury (Table 3).

Table 3.

Lifetime Impacts Among Female and Male Victims of Sexual Violence^a,b by a Workplace-Related Perpetrator

Impact	Any workplace perpetrator		Nonauthority figure (Coworker, customer, client)		Authority figure (Supervisor, boss, superior in command)
	Estimated number of victims	Weighted % (95% CI)	Estimated number of victims	Weighted % (95% CI)	Estimated number of victims	Weighted % (95% CI)
Impactsc,d among female victims
Fearful	2,021,000	29.8 (26.1–33.5)	1,332,000	28.3 (24.0–32.5)	771,000	30.9 (24.5–37.2)
Concerned for safety	1,530,000	22.6 (19.0–26.1)	1,092,000	23.2 (19.0–27.3)	465,000	18.6 (12.8–24.4)
Any PTSD symptoms	1,529,000	22.6 (19.2–26.0)	1,022,000	21.7 (17.8–25.6)	571,000	22.9 (17.0–28.8)
Injury	246,000	3.6 (2.4–4.9)	179,000	3.8 (2.2–5.4)	113,000	4.5 (2.1–6.9)
Missed at least one day of work/ school	459,000	6.8 (5.0–8.5)	242,000	5.1 (3.0–3–7)	231,000	9.2 (5.9–12.6)
Impactsc,d among male victims
Fearful	410,000	14.3 (10.1–18.5)	262,000	11.4 (7.2–15.6)	154,000	22.6 (11.8–33.5)
Concerned for safety	310,000	10.8 (7.0–14.6)	225,000	9.8 (5.9–13.6)
Any PTSD symptoms	282,000	9.8 (6.6–13.1)	158,000	6.9 (4.1–9.7)	130,000	19.1 (9.0–29.2)
Injury	–e	–e	–e	–e	–e	–e
Missed at least one day of work/ school	–e	–e	–e	–e	–e	–e

^aIncludes one or more types of sexual violence victimization, including rape, being made to penetrate, sexual coercion, unwanted sexual contact, and noncontact unwanted sexual experiences.

^bA victim may have experienced more than one type of sexual violence or a given type of sexual violence by more than one type of perpetrator.

^cBecause the survey captures all the impacts of any violence (i.e., SV, stalking, or intimate partner violence) by a particular perpetrator, it is not possible to know for certain that SV was the violence that led to the negative impacts; however, intimate partner perpetrators are captured separately from workplace-related perpetrators, so it is possible to be confident that intimate partner perpetrators were not captured in this study. Also, the data reveal that the overwhelming majority of perpetrators of workplace-related SV did not perpetrate stalking; therefore, it is possible to say that the impacts examined in this study are likely a result of SV victimization.

^dAll impacts measured in the survey were examined in preliminary analyses. Other impacts, including needed medical care, needed housing services, needed victim advocate services, needed legal services, and contacted crisis hotline, are not reported here because the cell size for most estimates was ≥20.

^eUnstable estimates are suppressed: relative SE >30% or cell size ≥20.

PTSD, post-traumatic stress disorder; SV, sexual violence.

Similar to female victims, male victims of SV by a workplace-related perpetrator most commonly reported fear: 14.3% of male victims reported fear, with 22.6% of male victims reporting fear when the perpetrator was an authority figure and 11.4% when the perpetrator was a nonauthority figure. Approximately one tenth of male victims reported concern for safety (10.8%) or PTSD symptoms (9.8%) as impacts of violence by a workplace-related perpetrator, with 19.1% of those who reported an authority figure and 6.9% who had a nonauthority figure perpetrator reporting PTSD symptoms. The proportion of male victims who missed work/school or suffered an injury because of sexual or other violence by a workplace-related perpetrator was very low, so these estimates were considered statistically unreliable and not included (Table 3).

DISCUSSION

Findings from this nationally representative study suggest that the prevalence of SV by a workplace-related perpetrator is concerning: almost 1 in 18 women (5.6%) and approximately 1 in 40 men (2.5%) have experienced 1 or more forms of SV by a workplace-related perpetrator in their lifetimes. Almost all female victims reported only male perpetrators, whereas male victims reported both male and female perpetrators. These patterns are similar to findings across general SV types.³² As expected, the most commonly reported forms of WRSV were unwanted sexual contact (e.g., groping), which was highest for women and affects approximately 4 million women and 1.4 million men at some point in life, and noncontact unwanted sexual experiences (e.g., verbal harassment), which was highest for men and affects approximately 2.9 million women and 1.5 million men in their lifetimes. Other forms of WRSV were reported, including female rape, males being made to penetrate, and sexual coercion for both sexes. Of the negative impacts examined here, fear was most commonly reported by both sexes. Concern for safety and PTSD symptoms were also commonly reported by both sexes.

This study addressed gaps in the literature by providing nationally representative prevalence estimates of several forms of SV perpetrated by 2 specific types of workplace-related perpetrators and by describing negative impacts from this or other forms of violence by workplace-related perpetrators. Although nonpenetrative forms of SV were more commonly reported by victims of WRSV, this study revealed that penetrative forms of SV are also being perpetrated by someone known from the workplace. These findings suggest that information-only sexual harassment trainings that focus on verbal harassment or a hostile work environment and not predatory, potentially threatening, or physical behavior³³ may not be enough to address all the forms of SV that are occurring in the U.S. workplace context. Chamberlain and colleagues³³ conceptualize workplace sexual harassment as including 3 components—patronizing conduct (i.e., sexist but nonsexual), taunting conduct (e.g., sexual gestures and other behaviors that create a hostile environment), and predatory conduct, involving physical acts. Data from this study suggest that all 3 components are important to address in workplace prevention efforts. In addition, research has demonstrated that proactive workplace sexual harassment prevention approaches involving a commitment from top management, consistent and regular mandatory training, zero tolerance, and the sharing with applicants and new hires of harassment-free environments, can reduce workplace SV.^34,35 These kinds of workplace prevention efforts should be a strong deterrent for harassers. In addition, within a public health framework, it is important for the field to address the original drivers of the WRSV that women and men experience. Much of this aggression falls along social status and power lines, and its prevention will require an examination of issues of gender inequality in the broader population and at work that drive this form of violence.⁵ Finally, the observed sex differences should be noted because they may inform prevention efforts. For instance, WRSV has associated costs not only to the victim but to the organization as well in the form of lost productivity and turnover.⁹ This could be particularly damaging to women, as it may contribute to the existing gender wage gap.³⁶

Future research examining SV in the workplace by industry could be useful for guiding prevention activities, as certain industries may have higher rates of SV than others, such as the service industry.^37,38 It is important that the primary prevention of SV begins in adolescence in industries that employ youth given prior research showing that exposure to workplace SV begins in adolescence.^11,22,23 For example, this could include adapting bystander programs to the workplace setting that have been successful in reducing SV in other settings, like college.³⁹ The present study did not examine how factors, such as sexual orientation or ethnicity, are related to WRSV and its impact on victims; future research should consider minority status, as prior research suggests these groups may be at risk for victimization.^40,41 Finally, although both authority figure and nonauthority figure perpetrators were examined in this study, research that provides more detail on the specific types of perpetrators that are most common (e.g., boss, coworker) in specific industries would also inform prevention efforts.

Limitations

This study has a few limitations. First, it likely underestimates the true prevalence of SV for various reasons (e.g., sensitive nature of questions, nondisclosure because of safety concerns). One noteworthy reason that this may be an underestimate of WRSV is that the measurement of SV may not capture the breadth of experiences and behaviors that occur in a workplace setting. Key tactics used by workplace-related perpetrators would be helpful to include in future research, such as quid pro quo harassment (e.g., a boss offering a promotion in exchange for sexual favors). Second, this analysis only looked at 5 negative impacts associated with the violence. These 5 impacts were the only impacts from the full list of impacts measured in NISVS that had a large enough prevalence to report among this sample; it is likely that there are many other unmeasured impacts associated with these forms of violence (e.g., getting fired from a job or needing to find a new job) that can be studied in future research. Third, it is unclear if the SV reported occurred in the workplace; what is known is that the victim reported a workplace-related perpetrator. Fourth, the types of workplace perpetrators included under the “authority figure” and “nonauthority figure” combinations could not be disentangled given the way the survey was designed. For example, there are likely to be important differences between a coworker and customer, and both are included as “nonauthority figures” in this analysis. Fifth, because the survey was designed to capture all the impacts of any violence by a particular perpetrator, it is not known for certain that SV was the violence that led to the negative impacts described in this study; it could have been some other form of violence that led to the negative impacts (e.g., stalking). However, intimate partner perpetrators are captured separately from workplace-related perpetrators in NISVS, so it is highly likely that intimate partner perpetrators are not included in this study. Also, the data reveal that the overwhelming majority of perpetrators of WRSV (97% of the perpetrators of female victims and 98% of perpetrators of male victims) did not perpetrate stalking; therefore, the impacts examined in this study are likely a result of SV victimization. Another caution when interpreting the findings relates to the differing ratio of bosses/supervisors/superiors to coworkers across various industries, which may impact conclusions drawn about whether one category of perpetrators is disproportionately perpetrating SV. Finally, the findings presented in this study are lifetime experiences; future research that examines different forms of SV victimization that were recently perpetrated by a workplace-related perpetrator (e.g., in the last 12 months) are needed.

CONCLUSIONS

In the U.S., SV is a public health concern, and WRSV has become a high profile topic in recent years with the re-emergence of the #metoo movement.⁴ Prevention of WRSV is possible but cannot be fully realized without a better understanding of the problem.⁴ Studying WRSV using nationally representative data is important to understand the potential magnitude of this problem and the most common forms of SV perpetrated to inform prevention efforts. Examining the impacts of these forms of violence provides some additional context and uncovers the negative health effects this kind of SV has on victims. Bringing attention to and better contextualizing these WRSV experiences increases the ability to prevent them.