The Prevalence and Psychological Cost of Interpersonal Violence in Graduate and Law School

Abstract

There is a rich research literature linking interpersonal violence to mental health disorders among undergraduate students. However, scholars know less about the prevalence and consequences of victimization among students enrolled in postbaccalaureate programs. Graduate and law students are uniquely vulnerable in their dependence on programs for financial support and career advancement, and they are more isolated than undergraduates. We explore the experiences of sexual harassment, sexual assault, and stalking in a sample of 1,149 male and female master’s, doctoral, and law students at a southeastern public university. First, the current study estimates prevalence rates of sexual harassment, sexual assault, and stalking across participant characteristics such as gender, race, sexual orientation, and degree program. We find higher rates of sexual harassment among lesbian, gay, bisexual, transgender, queer, intersex, or asexual (LGBTQ+) women and multiracial students and higher rates of coercion and stalking among women. Second, following a longstanding literature on experiences with violence among undergraduate women, we demonstrate the effects of interpersonal violence on depression and anxiety. A series of multivariate analysis of covariance (MANCOVA) suggests a heightened risk of mental health disturbances among law students, women, and LGBTQ+ students. Importantly, interpersonal violence partially explains these heightened levels of depression and anxiety in these groups. Given increasing rates of enrollment in postbaccalaureate programs, our findings suggest an unfulfilled need for the assessment, prevention, and treatment of individuals regarding the experiences of interpersonal violence and mental illness among graduate and law students. Student services should develop education and prevention programming for students and faculty to alleviate the occurrence and consequences of victimization while being mindful of the unique experiences of sexually and racially minoritized students.

Keywords

sexual assault sexual harassment stalking mental health and violence violence against LGBT

Research proliferates on undergraduates’ encounters with interpersonal violence and how these experiences affect well-being. For instance, a recent study on sexual harassment of college freshmen noted that female, White, and sexual minority students were at greater risk of chronic sexual harassment; chronic sexual harassment, in turn, is associated with higher levels of anxiety, depression, and substance abuse (McGinley, Wolff, Rospenda, Liu, & Richman, 2016). Wood, Hoefer, Kammer-Kerwick, Parra-Cardona, and Busch-Armendariz (2018) found that female students, younger undergraduate students, lesbian, gay, bisexual, transgender, queer (LGBTQ+) students, and non-Latinx White students were more likely to be victims of sexual harassment. Wood et al. engaged an intersectional framework seeking to understand whether race/ethnicity’s intersection with other identity categories put students of color at more risk. However, they do not find evidence for that hypothesis.

Similarly, female and LGBTQ+ undergraduate students are at an elevated risk of sexual assault; lesbian and bisexual women also report higher rates of anxiety, depression, and suicidal ideation than heterosexual women (Cantor et al., 2015; Eisenberg, Lust, Mathiason, & Porta, 2017; Kerr, Santurri, & Peters, 2013; Krebs, Lindquist, Warner, Fisher, & Martin, 2007). Unwanted, nonconsensual, and assaultive sexual experiences are consistently linked to affective disorders, substance abuse, and physical health problems in both the general population and among college students (Black et al., 2011; Harris & Valentiner, 2002; Hassija & Gray, 2013; Miranda, Meyerson, Long, Marx, & Simpson, 2002). There is less research, however, on the risk factors and consequences of sexual victimization among graduate and law students—individuals who attend the same institutions as undergraduates while facing distinct challenges.

Public discourse has begun to shed light on the harassment and assault of graduate and law students, such as in high-profile cases covered in The Chronicle of Higher Education, a major Internet and print news outlet for students, faculty, staff, and administrators (Brown, 2015; Mangan, 2015; Wilson, 2012). However, to date, broad empirical examination of interpersonal violence among graduate students is limited. One recent example is the National Academy of Science Report on Harassment of Women in Science, Technology, Engineering, and Mathematics (STEM) fields (Johnson, Widnall, & Benya, 2018). In a survey administered to Penn State Students, the authors found that 40% of female graduate students (across degrees) experienced harassment in the form of sexist comments and 18% experienced crude behavior. Another survey included in the report shows that Science, Engineering, and Medicine (SEM) students had higher rates of harassment than those in non-SEM fields.

A host of small-scale studies have examined sexual harassment of graduate or medical students in specific fields, although many of these studies were conducted almost 30 years ago or are limited in sample size and diversity (e.g., McKinney, Olson, & Satterfield, 1988; Miranda et al., 2002; Mohipp & Senn, 2008; Schneider, 1987). Several national-level, large-scale studies of sexual victimization and stalking in higher education do include graduate and professional students in their samples, but examination of these subgroups is limited (Cantor et al., 2015; Fisher, Cullen, & Turner, 2000). One recent contribution is McMahon, O’Connor, and Seabrook’s (2021) study, which compares undergraduate and graduate student experiences with sexual violence. However, importantly, this study focuses on disclosure and awareness of campus resources and does not distinguish between personal and educational characteristics of these postbaccalaureate participants.

Although graduate students are older and more likely to be married than undergraduates, two protective factors against victimization, they may be uniquely vulnerable because they generally work more closely with faculty and other graduate students (De Coster, Estes, & Mueller, 1999; Mohipp & Senn, 2008; Mustaine & Tewksbury, 2002). Graduate students are more isolated from the campus community and dependent on their programs and professors not only for grades (like undergraduates) but also for important opportunities in research, professionalization, and job prospects (McKinney, Olson, & Satterfield, 1988; McMahon et al., 2021; Schneider, 1987). They also hold dual roles as teachers and students, putting them at risk of harassment from both professors and students they teach, and they experience harassment and victimization while conducting research in the field (Hanson & Richards, 2017; Mohipp & Senn, 2008). Importantly, they are also less knowledgeable about and confident in seeking campus resources (McMahon et al., 2021).

Despite the importance of these studies for establishing patterns among college students and graduate students in specific programs, studies have not examined differences among graduate students. Additional research in this area is particularly warranted given that two of the studies that do compare graduate and undergraduate students find higher rates among the former group (Clodfelter, Turner, Hartman, & Kuhns, 2010; Cortina, Swan, Fitzgerald, & Waldo, 1998). The first research question of the current study is as follows:

Research Question 1: What is the prevalence of sexual harassment, sexual assault, and stalking among master’s, doctoral, and law students at a large, public, research-oriented university?

We also push this literature forward by employing an empirical understanding of intersectionality. Intersectional lenses are crucial to the understanding of victimization and crime more generally (Burgess-Proctor, 2006; Potter, 2013). Treating race, gender, or sexual orientation as uniform groups inadequately captures people’s lived experience. Following Crenshaw’s (1991) conceptualization of intersectionality as inequalities that are more than the sum of their parts, we contend that identities come together and are tied to structural disadvantage, situating people in particular ways. It is also crucial to restate that intersectionality is not just about identities; studies who reproduce this identities-only framework risk losing the very conceptual and theoretical power inherent in an intersectional lens. That is, intersectionality is about identities only to the extent that identities are attached to structurally and institutionally produced privileges and marginalizations.

Scholars have examined the intersections of race, gender, sexual orientation, citizenship status, and the like to theorize about and empirically study victimization and harassment in various domains and institutional settings, such as domestic violence, workplace harassment, sexual victimization, and stalking (Armstrong, Gleckman-Krut, & Johnson, 2018; Best, Edelman, Krieger, & Eliason, 2011; Bubar, 2014; Button & Worthen, 2014; Hernandez, 2000; Resko, 2007; Tester, 2008). Previous research has found heightened risk of various forms of victimization among groups such as African American or LGBTQ+ women (Andersen, Hughes, Zou, & Wilsnack, 2014; Crenshaw, 1991; Dennis, 1996). However, much of the quantitative research on interpersonal violence treats women as a monolithic group, and men are often neglected altogether. Thus, we ask:

Research Question 2: How do intersections of race, gender, and sexual orientation inform victimization risk in graduate and law school?

There is increasing dialogue in various news media, professional organizations, and websites about mental illness and substance abuse in graduate and law school (Djokic & Lounis, 2014; Louros, 2016; Rothstein, 2007; Schuman, 2014). Thus, there appears to be acknowledgment of both a harassment culture in academic departments and alcohol and psychological issues among graduate and law students. Although it has long been acknowledged that sexual harassment and discrimination have a psychological toll (e.g., Lenhart, 2004; Shrier, 1996), research has not fully explored the empirical connection between these experiences and the mental health of postbaccalaureate students to the extent that has been done in undergraduate populations. Two articles—one nearly 20 years old and one from 2016—demonstrate the necessity for exploring this relationship. Van Roosmalen and McDaniel (1999) connect sexual harassment to loss of self-esteem, negative emotion, and depression in a sample of undergraduate and graduate women. More recently, Rosenthal, Smidt, and Freyd (2016) found a significant relationship between sexual harassment and posttraumatic stress symptomatology in a sample of 324 female master’s, doctoral, and law students.

We update and build upon this literature by examining the effects of sexual harassment (hostile environment and unwanted contact), sexual assault (rape and coercion), and stalking on depression and anxiety in a sample of graduate and law students. We connect the literatures on victimization and mental health in this understudied population while contributing an intersectional approach. We explore:

Research Question 3: How do harassment, assault, and stalking relate to depression and anxiety among graduate and law students?

Method

Researchers received approval from the institutional review board (IRB) and the registrar to access student email addresses for students at a large, public, southeastern university (without participation of the administration). Only students enrolled in graduate and law school were emailed for participation in the “Graduate and Law Student Study” in April 2016 (N = 8,958). Participants provided informed consent by reading the consent document online and clicking “I agree.”

Upon survey completion, participants were automatically entered into a lottery for incentives ranging from US$10 to US$50, but students could opt into the lottery without participation. Sixty-six percent (n = 5,917) of students opened the email, 1,277 started the survey, and 1,149 completed the survey. A survey was considered complete if they submitted responses on their demographics, mental health, and experiences with interpersonal violence in their graduate or law program. Participants were excluded if they were not enrolled in a master’s, doctoral, or law degree program (n = 5); if they did not complete questions about sexual harassment, sexual assault, or stalking (n = 81); or if they did not complete measures of mental health (n = 42). White, non-Hispanic participants were more likely to complete the survey than racially minoritized participants (p < .05). In the end, 13% of students emailed, and 19% of students who opened the recruitment email are represented here (N = 1,149).

Although all enrolled postbaccalaureate students at this university were equally invited to participate in this study, selection bias is always a concern. Using data from the National Center for Education Statistics from the Department of Education, we compare our sample with the demographics of graduate students enrolled at this institution at this time. In comparison with the sampling frame, women are slightly more represented than males, making up 69% of this sample but only 60% of graduate and law students at this university during the same period (women also made up 60% of graduate and professional students nationally during this time). This may be because the consent form mentions sexual harassment and victimization, which may have compelled more women than men to participate. Black students are slightly underrepresented in our sample (7% of the sample vs. 10% of the sampling frame), and Asian students are overrepresented (11% vs. 5%). The number of Hispanic students (3%) and multiracial students (2%) at the university is closely proportional to our sample.

Gender and race/ethnicity differences in our sample are in line with decades of survey research in which women and White students are more likely to respond to surveys (Porter & Whitcomb, 2005). Our sample also matches the sampling frame closely in age. About half of graduate students enrolled at this time at this university were aged 25 to 35 (55% of our sample), 30% were 18 to 24 (31% in our sample), and 16% were 35 or older (14% in our sample).

Sexual Harassment

Participants were first presented with a definition of sexual harassment. “Sexual harassment is unwanted and unwelcome sexual behavior which interferes with your life, which does not include behaviors that you like or want (for example, wanted kissing, touching, or flirting” (Hill & Silva, 2005, p. 6). They were then asked, “Since starting graduate/professional/law school, has someone done the following during or after college-related events or activities?” (e.g., such as in class, on campus, and during school- or department-sponsored events [e.g., social events, study groups, meetings]). They were then presented with a list of 15 behaviors. This definition and methodology follows Hill and Silva’s (2005) report for the American Association of University Women (AAUW) titled “Drawing the Line: Sexual Harassment on Campus,” as they reflect both the experiences of sexual harassment among college students and the laws and policies implemented on college campuses. The alpha for this scale is .79.

The potential responses were “no, never,” “yes, once,” and “yes, more than once” for each of the 15 items. Participants were coded 1 for hostile environment if they experienced one or more of seven items that involve behaviors that while they may intimidate or be seen as sexually aggressive, they do not involve physical contact (e.g., made sexual comments, jokes, gestures, or looks) and 0 if they experienced none of these behaviors. Participants were coded 1 for sexual contact if they experienced one or more of seven different items that involve physical, sexualized contact (e.g., touched, grabbed, or pinched in a sexual way; forced to do something sexual). Because quid pro quo harassment (being “asked to do something sexual in exchange for something, such as a better grade”) was so rare (<1%), this form of harassment was not analyzed.

Rape and Coercion

Coercion, despite its sexually abusive nature, is rarely seen as “rape” by victim/survivors (e.g., Boyle & McKinzie, 2015). Rape and coercion are separated as they are related yet conceptually different experiences and thus are often examined separately (e.g., Krebs et al., 2016). Unlike our sexual harassment instruments, participants were not presented with definitions of “rape,” “coercion,” or “stalking,” and those words were not used explicitly. Participants were asked to indicate whether “anyone has done the following since you started post-undergraduate study (graduate, law, or professional school).” They were then presented with a list of behaviors defined as sexual assault by the National Intimate Partner and Sexual Violence Survey (Black et al., 2011).

The definition of rape contains multiple parts: physical force and incapacitation. Participants were asked whether anyone used “physical force or threats to physically harm you” to “make you have” vaginal sex, anal sex, or oral sex. They were also asked whether force was used to penetrate their vagina or anus with fingers or an object and whether someone tried to use force to have sex with them but sex did not happen. They were also asked whether, “when you were drunk, high, drugged, or passed out and unable to consent,” someone had vaginal, anal, or oral sex with them. These acts were defined: “vaginal sex, meaning that a man or boy put his penis in your vagina”; “anal sex, meaning that a man or boy put his penis in your anus”; “perform oral sex, meaning that a man or boy put your mouth on their penis or anus”; and “receive oral sex, meaning they put their mouth on your vagina or anus.” Participants who self-identified as men at the start of the survey were asked questions that were adapted accordingly (e.g., “by vaginal sex, we mean that a woman or girl made you put your penis in her vagina”). In order to be inclusive, participants who self-identified as transgender or other non-cisgender identities answered questions that contained both options listed above, as biological sex was not ascertained.

For each item, their response options were “Yes” and “No.” These forms of forcible and incapacitated rape were collapsed into rape. Participants are coded 1 if they answered affirmatively to any of these questions and 0 if they answered no to every question.

Participants were also asked about coercion. They are coded 1 for coercion if they indicated that someone had vaginal, oral, or anal sex with them after being pressured through three different tactics: “telling you lies, making promises about the future they knew were untrue, threatening to end your relationship, or threatening to spread rumors about you”; “wearing you down by repeatedly asking for sex or showing they were unhappy”; and “using their authority over you, for example, your boss or your teacher.” Their response options were “Yes” and “No.” Participants were coded 0 for coercion if they did not have any of these experiences and 1 if they had one or more of these experiences since starting graduate/law/professional school.

Stalking

Stalking was also measured using the National Intimate Partner and Sexual Violence Survey (Black et al., 2011, p. 23). Participants were asked, “since starting post-undergraduate study (graduate, law, or professional school, etc.), has anyone . . .” and then they were presented with eight behaviors, such as someone “watched or followed you from a distance or spied on you with a listening device, camera, or GPS (global positioning system)”; “sneaked into your home or car and did things to scare you by letting you know they had been there”; and “left you cards, letters, flowers, or presents when they knew you didn’t want them.” Their options were “No, never”; “Yes, once”; and “Yes, more than once.” They were coded 1 for stalking if they experienced any of these behaviors once or more than once and 0 if they responded “No, never” to all of the eight items.

We did not examine other forms of interpersonal violence, such as intimate partner abuse, because we are interested in the context of harassment and assault within graduate and law school rather than in relationships more broadly.

Mental Health

The depression (α = .94) and anxiety (α = 0.90) subscales of the Depression Anxiety Stress Scales (DASS) were used to assess mental health during the past week (Lovibond & Lovibond, 1995). Twenty-eight items are rated from 0 (did not apply) to 3 (applied very much or most of the time) and were used to create the two summative scales. The mean for each form of mental health is listed for each subgroup in Table 2.

Participant Characteristics

Participants completed the sentence “I identify as . . .” (Table 1). Most students identified as women (69%) or men (31%), while eight selected transgender, gender fluid, or provided alternative labels such as genderqueer. Participants identified as heterosexual or straight (90%), gay or lesbian (4%), or bisexual (4%). Less than 2% total selected asexual or other, with the most common write-in being queer. Due to issues of statistical power, lesbian, bisexual, and other orientations are collapsed into LGBTQ+ in multivariate analyses (10%). There are certainly important differences in the experiences of members of these subgroups, which we discuss where we are able to.

Table 1.

Characteristics and Interpersonal Violence in the Sample.

Demographics	n (%)	Hostile Environment	Sexual Contact	Rape	Coercion	Stalking
Demographics	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)
Total sample	1,149	380 (33)	125 (11)	78 (7)	75 (7)	172 (15)
Prevalence rates
Men	361 (31)	126 (35)	35 (10)	20 (6)	11 (3)	41 (11)
Women	788 (69)	254 (32)	90 (11)	58 (7)	64 (8)	131 (17)
χ²		0.7971	0.7607	1.296	10.450**	5.396*
Cramer’s V					0.0954	0.0685
Heterosexual	1,033 (90)	322 (31)	107 (10)	72 (7)	67 (6)	150 (15)
Gay or lesbian	51 (4)	24 (47)	6 (12)	2 (4)	1 (2)	7 (14)
Bisexual	48 (4)	26 (54)	11 (23)	3 (6)	6 (13)	10 (20)
χ²		15.872***	7.453*	0.7365	4.546	1.498
Cramer’s V		0.1184	0.0811
White non-Hispanic	878 (76)	307 (36)	96 (11)	51 (6)	57 (7)	129 (15)
Black or African American	75 (7)	15 (21)	5 (7)	8 (11)	8 (11)	6 (9)
Asian	121 (11)	20 (17)	9 (8)	8 (7)	2 (2)	19 (16)
Hispanic or Latinx	51 (4)	18 (37)	8 (16)	6 (12)	5 (10)	7 (14)
Multiracial	43 (3)	18 (42)	5 (12)	4 (9)	3 (7)	11 (26)
χ²		22.519***	3.904	6.166	8.326	6.147
Cramer’s V		0.1405
Master’s student	420 (37)	107 (25)	29 (7)	23 (5)	19 (5)	56 (13)
Doctoral student	632 (55)	228 (36)	79 (13)	48 (8)	47 (7)	109 (17)
Law student	87 (8)	45 (46)	17 (18)	7 (7)	9 (9)	7 (7)
χ²		21.299***	12.975**	1.82	4.823	8.039*
Cramer’s V		0.1362	0.1063			0.0836
Intersectional category
Heterosexual White women	544 (47)	177 (33)	56 (10)	38 (7)	44 (8)	89 (16)
Heterosexual women of color	175 (15)	41 (23)	20 (11)	16 (9)	13 (7)	27 (15)
Heterosexual White men	244 (21)	93 (38)	26 (11)	11 (5)	8 (3)	30 (12)
Heterosexual men of color	74 (6)	12 (16)	5 (7)	8 (11)	2 (3)	4 (5)
LGBTQA+ women	69 (6)	36 (52)	14 (20)	4 (6)	7 (10)	15 (22)
LGBTQA+ men	43 (4)	21 (49)	4 (9)	1 (2)	1 (2)	7 (16)
χ²		35.928***	7.969	6.924	11.124*	10.087
Cramer’s V		0.1768			0.0984

Note. LGBTQA = lesbian, gay, bisexual, transgender, queer, intersex, or asexual.

p < .05. **p < .01. ***p < .001.

In bivariate analyses, we also examine experiences at intersections of race, gender, and sexual orientation. The largest intersectional group was White heterosexual women (47%), followed by White heterosexual men (21%), heterosexual women of color (15%), heterosexual men of color (6%), LGBTQ+ women (6%), and LGBTQ+ men (4%) (Table 1). Students are categorized as a man or woman of color if they are not solely White non-Hispanic.

In our analyses, we control for a number of other participant characteristics. The average age was almost 29 and most were single and never married (69%) or were married or in a domestic partnership (27%). The largest group consists of doctoral students (55%), followed by master’s (37%) and law (8%) students. Due to the lower number of law and master’s students, doctoral students are compared with both other groups in multivariate analysis. On average, participants were in their program for 2.6 years. About a quarter (26%) were first-generation college students (neither parent/guardian graduated college with a 4-year degree) and 12% are coded as international students, having spent most of their lives outside the United States.

Analytic Strategy

First, we establish differences in victimization (Table 1). For descriptive purposes, we report the percentage of each group that has experienced each form of victimization and their mean score on depression and anxiety measures. As victimization variables are categorical, chi-squares are reported along with Cramer’s V. We then assess bivariate differences in mental health (Table 2). Depression and anxiety are continuous, so we estimate a series of bivariate analyses of variance (ANOVAs) and report Eta-squared (η²), a measure of effect size.

Table 2.

Levels of Depression and Anxiety in the Sample.

	Depression	Anxiety
	M (SD)	M (SD)
Total sample	6.342 (7.339)	4.563 (5.847)
Skewness	1.904	2.295
Kurtosis	6.788	9.572
Men	6.584 (7.369)	4.050 (5.289)
Women	6.231 (7.328)	4.798 (6.074)
η²	.001	0.002
Heterosexual	6.061 (7.150)	4.412 (5.787)
Gay or lesbian	7.275 (7.523)	4.471 (6.031)
Bisexual	9.583 (8.545)	6.792 (5.929)
η²	.011**	.007*
White non-Hispanic	6.135 (6.970)	4.453 (5.584)
Black or African American	4.671 (7.304)	3.786 (6.169)
Asian	7.093 (7.798)	3.915 (6.905)
Hispanic or Latinx	8.184 (8.698)	5.816 (6.405)
Multiracial	9.349 (10.461)	5.767 (6.668)
η²	.005*	0.001
Master’s student	6.336 (7.243)	4.729 (5.919)
Doctoral student	5.973 (6.861)	4.342 (5.649)
Law student	8.773 (9.940)	5.289 (6.721)
η²	.007*	0.002
Intersectional category
Heterosexual White women	5.765 (6.806)	4.526 (5.721)
Heterosexual women of color	6.366 (7.894)	4.846 (6.838)
Heterosexual White men	6.143 (6.830)	3.758 (4.924)
Heterosexual men of color	7.095 (8.712)	4.676 (6.093)
LGBTQA+ women	9.565 (8.887)	6.826 (6.421)
LGBTQA+ men	8.209 (7.701)	4.628 (5.778)
η²	.018***	.011*

Note. LGBTQA = lesbian, gay, bisexual, transgender, queer, intersex, or asexual.

p < .05. **p < .01. ***p < .001.

Second, we predict victimization in multivariate analyses with a series of control variables (Table 3). We use multivariate analysis of covariance (MANCOVA), which is a common analysis employed when there are multiple correlated dependent variables and multiple predictors. Here, we predict our five interpersonal violence measures in a single model: hostile environment, sexual contact, rape, coercion, and stalking. We report unstandardized beta coefficients, which are produced by the mvreg command in STATA.

Table 3.

Multivariate Models Predicting Interpersonal Violence.

Demographics	Hostile Environment		Sexual Contact		Rape		Coercion		Stalking
Demographics	b	SE	b	SE	b	SE	b	SE	b	SE
Women	−0.046	0.029	0.011	0.020	0.013	0.016	0.049	0.016**	0.049	0.023*
Men (ref. group)
LGBTQ+	0.183	0.046***	0.057	0.031	−0.030	0.025	0.004	0.025	0.046	0.036
Heterosexual (ref. group)
Black, Hispanic, or multiracial	−0.024	0.039	0.015	0.027	0.053	0.021*	0.026	0.021	−0.011	0.030
Asian	−0.121	0.057*	−0.012	0.039	−0.001	0.031	−0.063	0.031*	−0.006	0.044
White non-Hispanic (ref. group)
Doctoral students	0.081	0.029**	0.038	0.020	0.024	0.016	0.018	0.016	0.055	0.023*
Years in the program	0.009	0.007	0.004	0.005	0.002	0.004	0.007	0.004	0.007	0.006
First-generation students	−0.041	0.031	−0.021	0.021	0.004	0.017	−0.001	0.017	−0.014	0.024
International	−0.125	0.053*	−0.043	0.036	0.003	0.029	0.016	0.028	0.014	0.041
Age	−0.006	0.002**	−0.003	0.002*	−0.001	0.001	−0.001	0.001	−0.001	0.002
Never married	0.094	0.035**	0.010	0.023	0.034	0.019	0.033	0.019	0.088	0.027**
Constant	0.432	0.083	0.159	0.057	0.048	0.046	0.003	0.045	0.032	0.065
RMSE	0.457		0.310		0.251		0.245		0.354
R ²	.067		.018		.016		.023		.027
F	8.214***		2.042*		1.788		2.727**		3.163***

Note. b = beta coefficient; LGBTQA = lesbian, gay, bisexual, transgender, queer, intersex, or asexual; RMSE = root mean square error.

p < .05. **p < .01. ***p < .001.

Third, we use MANCOVA to predict depression and anxiety (Table 4). We examine the effects of participant characteristics and violence on depression and anxiety simultaneously. Because both depression and anxiety are positively skewed with heavy tails, we use the STATA postestimation command estat imtest after multivariate analyses, which demonstrated the presence of heteroskedasticity. The use of robust standard errors relaxes assumptions of independent and identical distributions of standard errors and are used herein (Allison, 1999).

Table 4.

Multivariate Models Predicting Mental Health.

Demographics	Depression		Anxiety
Demographics	b	RSE	b	RSE
Women	−0.581	0.473	0.522	0.353
Men (ref. group)
LGBTQ+	2.780	0.793***	1.492	0.581*
Heterosexual (ref. group)
Black, Hispanic, or multiracial	0.738	0.692	0.355	0.524
Asian	1.375	0.912	0.809	0.799
White non-Hispanic (ref. group)
Doctoral students	−1.232	0.478*	−0.714	0.358*
Years in the program	0.079	0.120	−0.026	0.085
First-generation students	0.025	0.495	−0.311	0.396
International students	−0.274	0.805	−0.074	0.666
Age	0.008	0.031	0.007	0.025
Never married	0.545	0.524	0.113	0.406
Hostile environment	0.000	0.488	0.262	0.407
Sexual contact	1.612	0.827	1.225	0.658
Rape	−0.175	0.860	0.487	0.827
Coercion	1.560	0.954	2.490	0.955**
Stalking	2.644	0.726***	1.736	0.561**
Constant	5.458	1.189	3.510	0.982
RMSE	7.187		5.724
R ²	.054		.054
F	4.279***		4.325***

Note. b = beta coefficient; RSE = robust standard error; LGBTQA = lesbian, gay, bisexual, transgender, queer, intersex, or asexual; RMSE = root mean square error.

p < .05. **p < .01. ***p < .001.

To further demonstrate how victimization is related to mental health, we estimate a series of Sobel–Goodman mediation tests, which calculate the direct and indirect effects of variables while controlling for relevant other variables. Here, we determine whether the effects of participants’ characteristics (gender, sexual orientation, race/ethnicity, and program of study) on mental health are partially explained by their victimization experiences. Thus, we establish—albeit cross-sectionally—the relationship between participant characteristics, interpersonal violence risk, and mental health.

Before continuing with results, we make a note of statistical power. Ideally, analyses would go beyond dichotomizing important, complex variables like sexual orientation and race/ethnicity. In our bivariate analyses, we consider differences between, for instance, lesbian and bisexual participants and various intersections between race, gender, and sexual orientation. Detecting a medium effect size (.15) with the statistical power level of .8 warrants at least 42 participants to populate each subgroup (Cohen, 1988; Soper, 2006-2018). The smallest groups are multiracial participants and LGBTQ+ men (n = 43), suggesting we have adequate power to detect differences. Although some scholars argue for a cell size of at least 50, they contend that these are heuristics, not definite cutoffs, particularly when effect sizes are unknown, and analyses are exploratory (Simmons, Nelson, & Simonsohn, 2013).

Our final multivariate models contain 15 predictors, which limits our statistical power (Soper, 2006-2018). Thus, in multivariate analyses, we dichotomize sexual orientation (heterosexual, LGBTQ+) and collapse race/ethnicity into three groups (White, non-Hispanic; Asian; Black, Hispanic/Latinx, or multiracial). We do not display intersectional differences in multivariate results, although we hope that these exploratory analyses motivate future research that uses intersectional approaches.

Results

Sexual Harassment

Table 1 displays prevalence rates of interpersonal violence across participant characteristics. For each characteristic and form of victimization, the chi-square statistic is reported to determine the significance of these differences. Overall, a third of participants (33%) experienced a hostile environment and more than one in 10 (11%) experienced some form of unwanted sexual contact in their graduate or law program. By far, the most common form was sexual comments, jokes, gestures, and looks (30%), followed by being brushed up against and being touched, grabbed, or pinched in a sexual way (both about 8%). Nearly half of law students experienced one or more elements of a hostile environment (46%), followed by doctoral (36%) and master’s (25%) students (χ² = 21.299, p < .001).

A statistically equivalent number of men and women reported sexual harassment—although the details of their experiences varied. For instance, men were significantly more likely than women to report homophobic harassment (being called “gay,” “lesbian,” or a homophobic name) and more likely to be flashed or mooned (p < .05). LGBTQ+ students were more likely to experience not only homophobic harassment but also a range of other forms (e.g., targeted with sexual rumors, forced to do something sexual). Nearly half of gay or lesbian (47%) and bisexual students (54%) reported a hostile environment, a significantly higher rate than reported by heterosexual students (31%) (χ² = 15.872, p < .001). Almost one in four bisexual students (23%) reported unwanted sexual contact in their program, which is higher than heterosexual (10%) and gay or lesbian students (12%) (χ² = 7.453, p < .05).

Race/ethnicity was also a predictive factor in experiences with a hostile environment (χ² = 22.519, p < .001). Asian students reported the lowest rates (17%), followed by Black or African American students (21%). About 36% of both White, non-Hispanic and Hispanic/Latinx reported a hostile environment, and the highest rate was found among students who are multiracial (42%). These patterns did not persist, significantly, for unwanted sexual contact.

Among intersectional groups, the highest prevalence of a hostile environment was among LGBTQ+ women (52%), followed by LGBTQ+ men (49%), White heterosexual men (38%), White heterosexual women (33%), women of color (23%), and men of color (16%). These differences are statistically significant (χ² = 35.928, p < .001). LGBTQ+ women reported the highest rates of sexual contact (20%), twice as high as White heterosexual women (10%)—a finding that is partially driven by bisexual women. Because patterns for hostile environment and sexual contact diverge in some ways across race, gender, and sexual orientation, it is important to study both forms.

Rape, Coercion, and Stalking

Eleven percent of students experienced at least one form of sexual assault since starting their graduate or law school program. More specifically, 7% experienced some form of coercion and 7% experienced some form of incapacitated or forcible rape (Table 1). Coercive use of authority was the rarest form of assault (<1%). Coercion through lies or nonphysical threats and forcible rape were equally common (about 3%). The most common forms of sexual assault were incapacitated rape (5%) and coercion through persistent pressure (5%). While there were great disparities in sexual harassment experiences based on degree program, fewer differences emerge for rape and coercion: While master’s students experienced each form less frequently, these differences do not reach statistical significance.

Twice as many women (8%) than men (3%) were pressured, lied to, or coerced through authority into unwanted sex (χ² = 5.396, p < .05). The highest rate of coercion is found among bisexual students (13%), a figure that is only significant when compared with both heterosexual and gay or lesbian students (χ² = 3.897, p < .05) (results available upon request). When examining intersections, it appears that LGBTQ+ women experienced coercion at higher rates than any other group (χ² = 11.124, p < .05). Ten percent of LGBTQ+ women reported coercion, as compared with one (2%) LGBTQ+ man, and 3% of heterosexual White men and men of color.

The most commonly experienced forms of stalking were unwanted messages via telephone, text, email, or websites like Facebook (9%). The next most common behavior was being left gifts when the person knew they were unwanted (3%); the least common was having one’s home or car snuck into (<1%). Doctoral students (17%) were the most likely to experience this (χ² = 8.039, p < .05), and women (17%) were about 60% more likely than men (11%) to experience at least one form of stalking while in graduate or law school (χ² = 5.396, p < .05). There were no differences in stalking based on race/ethnicity or sexual orientation.

Multivariate Results

Table 3 displays the results of a MANCOVA that compares the prevalence of interpersonal violence across participant characteristics while controlling for the number of years in their program, first-generation and international student status, age, and marital status.

There are still no significant gender differences for sexual harassment, and LGBTQ+ participants reported significantly higher levels of a hostile environment than heterosexual students (b = 0.183, p < .001). Women are still more likely than men to report coercion (b = 0.049, p < .01) and stalking (b = 0.049, p < .05). Asian students are less likely to experience a hostile environment (b = −0.121, p < .05) but also less likely to experience sexual coercion than White, non-Hispanic students (b = −0.063, p < .05).

Not previously detected in bivariate analyses, Black, Hispanic/Latinx, or multiracial students were more likely to experience rape while in graduate or law school than White, non-Hispanic students (b = 0.053, p < .05). While race/ethnicity is the only significant predictor of experiencing rape, the model F statistic only approaches significance (F = 1.788, p < .10). The removal of nonsignificant variables in various combinations maintains the significance of race while producing a significant F statistic (p < .05).

Other participant characteristics are worth mentioning here. International students (b = −0.125, p < .05) and older students (b = −0.006, p < .01) are less likely to report experiencing a hostile environment than students who grew up in the United States and younger students, respectively. Older students are also less likely to experience unwanted sexual contact (b = −0.003, p < .05), while single and never married students are more likely to experience a hostile environment (b = 0.094, p < .01) and stalking (b = 0.088, p < .01) than students who are married, divorced, or in a domestic partnership.

Differences by program

Doctoral students report higher levels of a hostile environment, even while controlling for the fact that their programs are longer than master’s or law programs (b = 0.081, p < .01). Doctoral students are also more likely to have experienced stalking in their program (b = 0.06, p < .05). In supplementary analyses, we alternate the reference group (available upon request). Compared with master’s students, both doctoral and law students are more likely to report a hostile environment and sexual contact. When doctoral students are the reference group, master’s students are more likely to report a hostile environment and sexual contact.

Depression and Anxiety

Table 2 shows the effect size (η²) produced by a series of bivariate ANOVAs between participant characteristics and depression and anxiety. Bisexual students report the highest rates of depression and anxiety in the sample (p < .05). Depression rates also varied significantly by race/ethnicity and degree program. Twenty-two percent of multiracial participants versus 14% of White, non-Hispanic and 11% of Black or African American participants meet the criteria for moderate or more severe levels of depression. Law students also report fairly high levels of depression (M = 8.773), a level that is significantly higher than master’s and doctoral students combined (p < .05). When looking at the combination of gender, race, and sexual orientation, LGBTQA+ women, followed by LGBTQA+ men, have the highest rates of depression (p < .001). LGBTQA+ women also have the highest rates of anxiety (M = 6.826), followed by heterosexual women of color (M = 4.846). Heterosexual White women had the lowest levels of depression, and heterosexual White men had the lowest scores of anxiety.

In multivariate analyses, we consider the role that interpersonal violence plays in producing mental health issues. Table 4 presents the results of a MANCOVA that compares rates of depression and anxiety among graduate and law students. Notably, stalking significantly and positively predicts both depression (b = 2.644, p < .001) and anxiety (b = 1.736, p < .01). Experiencing coercion is also associated with higher rates of anxiety (b = 2.490, p < .01). As for sexual harassment, unwanted sexual contact only approaches significance in both models when using robust standard errors, warranting further attention in future studies (p < .10).

Even in multivariate analyses, LGBTQ+ students continue to have significantly higher rates of depression and anxiety than heterosexual participants (b = 2.780, p < .001; b = 1.492, p < .05). When accounting for experiences with interpersonal violence and personal characteristics, doctoral students have significantly lower levels of depression (b = −1.232, p < .05) and anxiety (b = −0.714, p < .05). In alternative analyses (available upon request), law students are included instead of doctoral students, showing significantly higher levels of depression than master’s and doctorate seeking participants (b = 2.858, p < .001). While a larger comparison group of law students is needed to better detect these effects, there is evidence that they have poorer mental health than graduate students.

Our dichotomous measures of interpersonal violence—“victim” versus “nonvictim” in each category—do not account for the frequency of victimization or the various forms of each type of violence one might experience (e.g., forcible oral sex vs. incapacitated vaginal penetration). In supplementary analyses, we weight within-construct items by their factor loading and then sum into a single scale variable for each of the five forms of violence. These analyses do not produce different results in predicting mental health. Stalking is still the only measure that predicts depression, and coercion and stalking are the only measures that predict anxiety (available upon request).

Differences by program

To assess differences by degree program, we ran supplementary analyses in which we alternate which group is used as the reference group. These findings reinforce bivariate, descriptive findings in Table 2 and are available upon request. Master’s (b = −2.44, p < .01) and doctoral students (b = −3.21, p < .001) have lower levels of depression than law students (reference group). Correspondingly, when doctoral students are the reference group, we find law students have greater depression (b = 3.21, p < .001), but master’s students do not differ from doctoral students. Regardless of the reference group, program of study does not affect anxiety.

Mediation Effects

To ascertain whether interpersonal violence mediates the relationship between participant characteristics and mental health, a series of Sobel–Goodman analyses were conducted (available upon request). Some bivariate effects on mental illness can be partially explained by exposure to interpersonal violence. For instance, 18% of the effect of program of study on depression is explained by sexual harassment in the form of unwanted sexual contact. More specifically, law students’ higher rates of anxiety and depression are partially caused by experiencing sexual harassment and assault in their law program.

Interpersonal violence is key to understanding levels of psychological distress among women and LGBTQ+ participants. Even though sexual orientation is still significant in multivariate models, part of its effect (11%) on anxiety is mediated by exposure to a hostile environment (p < .05). Ten percent of the relationship between bisexuality and anxiety is explained by unwanted sexual contact (p < .05), and 9% is explained by experiencing a hostile environment.

As shown in Tables 2 and 3, the intersections of identities by race, gender, and sexual orientation are related to rates of both victimization and mental health. Almost a fourth (23%) of the effect of intersectional identity on anxiety is explained by exposure to a hostile environment (p < .05). LGBTQ+ women had the highest rates of anxiety and depression, which is partially explained by their greater exposure to a hostile environment and sexual contact, respectively (p < .05). Conversely, lower rates of depression and anxiety among heterosexual White men are partially explained by their limited exposure to sexual coercion (p < .05).

In sum, LGBTQ+ students’ higher rates of anxiety and depression—particularly bisexual students and LGBTQ+ women —and law students’ increased mental health concerns can be partially attributed to their experiences with interpersonal violence.

Discussion

In a research note published in 1988, McKinney and colleagues cited a host of studies conducted in the 1980s that found graduate student sexual harassment rates of between 20% and 35%. Their prevalence rate was within these bounds, as is ours. Thirty years later, we found 33% of graduate or law students have experienced a hostile environment, while one in 10 experienced sexual harassment in the form of unwelcome sexual contact. On average, 11% of postbaccalaureate students in this study experienced rape or coercion (or both), and one in seven experienced stalking, in the 2 to 3 years of their program. Despite much public discussion of this pressing and consequential problem, little research has theoretically or empirically sought to address harassment—and interpersonal violence more generally—in this population (Rosenthal et al., 2016).

This study is certainly not the first to claim there is a problem of sexual harassment in law school (Lerman, 2006; Torrey, Ries, & Spiliopoulos, 1997) or graduate school (Schneider, 1987). However, studies tend to be dated, to examine particular degree programs, or to not examine differences between students. Large-scale, methodologically sophisticated studies like the Association of American Universities (AAU) climate study (Cantor et al., 2015) provide prevalence measures for graduate students, but do not distinguish differences by degree program or other important characteristics. This dearth of knowledge limits our ability to understand who is more at risk and the consequences of interpersonal violence in this population. The current study attends to these gaps in the literature by exploring the victimization of master’s, doctoral, and law students.

In addition, we pay close attention to intersecting inequalities. This is paramount for at least three reasons. First, the category of woman (or man) is obviously nuanced by class, race, ethnicity, religion, ability, and sexual orientation. Treating men and women, gay and straight, White and non-White students as members of homogeneous groups impedes social scientific understanding of group similarities and differences (Crenshaw, 1991). Second, intersectionality’s focus on institutionally produced domination and power (that affects people disparately based on the many ways in which they are categorized) reminds us to look to institutions and policy for redress. Third, while race, gender, and so on are socially constructed, there are real material consequences of those categorizations and thus, we must know more about differences and similarities between and among groups and between and among categories. Although our ability to fully investigate the important ways in which institutionalized forms of inequality intersect to produce domination and oppression is limited, we discuss some key findings and suggestions for future research below.

Graduate and Law Programs

Quid pro quo sexual harassment and coercive use of authority in sexual assault—examples of egregious abuse of power often covered in news media—were quite rare. However, a third of students reported experiencing at least one element of a hostile environment and 11% experienced sexual harassment in the form of sexual contact. These findings varied by degree program. Nearly half of law students self-reported a hostile environment, and almost one in five experienced unwanted sexual contact in the context of their study. This is significantly higher than master’s students even while controlling for a number of characteristics, such as age, marital status, and years in their program.

Doctoral students were not far behind law students, also experiencing significantly higher levels of a hostile environment and sexual contact than master’s students. Doctoral students were also more than twice as likely as law students to experience stalking since starting graduate school. Stalking is the only form of interpersonal violence in which master’s students scored higher than law students. This calls into question the dynamics present and experiences that occur in academic departments where both masters and doctorates are granted while acknowledging that there may be aspects of pursuing a doctorate degree that raises the risk of sexual harassment.

Race/Ethnicity

Sexual harassment rates were also high among multiracial students, although small numbers of participants in these categories warrant caution in interpreting results. Our findings are somewhat contrary to nationally representative studies of sexual victimization, which find similar rates between White and Black participants or higher rates among White students (Black et al., 2011). Some research, however, shows that White students are only more at risk of incapacitated rape and assault (Krebs et al., 2007). Although race has been understudied in quantitative studies of sexual violence, Mohler-Kuo, Dowdall, Koss, and Wechsler (2004) and Gross, Winslett, Roberts, and Gohm (2006) found that White undergraduate females were actually less likely to experience forcible rape and emotional coercion than other groups, and Black women in particular. This is consistent with this study’s findings.

In future studies, careful attention should be paid to how sexualized stereotypes and cultural norms influence the targeting of some students by sexual offenders, as well as how barriers to reporting victimization and help-seeking might be heightened among students of particular groups (Cortina et al., 1998; Eisenberg et al., 2017; Kalof, Eby, Matheson, & Kroska, 2001; Tang, Yik, Cheung, Choi, & Au, 1996). This matter is particularly pressing because research shows that African Americans and Latinos are less likely than Whites to report sexual assault to authorities (Baumer & Lauritsen, 2010; Boyle & Walker, 2016; Tarling & Morris, 2010).

Gender and Sexual Orientation

Analyses of sexual harassment and assault in graduate and law programs demonstrate the need to both further study postbaccalaureate programs and examine multiple participant characteristics simultaneously. For example, women did not differ significantly from men in their rates of sexual harassment, unlike previous studies and the conventional knowledge on campus rape, which focuses on women (Cantor et al., 2015; Hill & Silva, 2005). Other findings are more consistent with previous research, such as the fact that LGBTQ+ students had the highest rates of sexual harassment. This finding, however, is largely driven by the experiences of bisexual students, who had higher rates of a hostile environment than heterosexual and gay or lesbian students and rates of sexual contact twice as high—which is consequential for mental health. LGBTQ+ men, on the contrary, had lower rates of sexual harassment and stalking than LGBTQ+ women, highlighting the need to understand how multifaceted identities relate to discrimination and oppression. Understanding how sexual orientation positions men and women differently in the specific context of graduate and law school is paramount for alleviating the consequences of these experiences.

Unlike many studies, women (as a uniform category) did not report higher rates of sexual harassment or rape than men (Cantor et al., 2015; Tjaden & Thoennes, 2000). However, they were significantly more likely to experience coercion and stalking. Seventeen percent of women and 11% of men self-reported at least one form of stalking, which is in line with gender-differentiated lifetime prevalence rates in a meta-analysis of 175 studies of stalking (Spitzberg & Cupach, 2007). Marital status is also a significant predictor of stalking. Single, never married students are significantly more likely to be targeted by stalking behavior such as being followed, threatened, or sent unwanted messages or gifts. It is unclear, however, whether these stalking incidents occurred at the hands of professors, students, staff, or someone else entirely unrelated to graduate study. It is crucial that research furthers understanding of risk factors and perpetration tactics of stalking, as stalking has a greater impact on depression and anxiety than any other form of interpersonal violence studied here.

Psychological Costs of Interpersonal Violence

The third research question aimed to explore the consequences of interpersonal violence among graduate and law students. Higher rates of depression can be found among multiracial students and law students, while women and LGBTQ+ students reported higher rates of anxiety. Of particular concern is the fact that we find that the average bisexual or multiracial participant approaches the cutoff for mild levels of depression (about 9.5): A score of 10 is the recommended cutoff score (Lovibond & Lovibond, 1995). Nearly a third of bisexual participants reach the cutoff score of moderate to extremely severe depression, as compared with 14% of heterosexual participants.

While controlling for participant characteristics like age and marital status, which are generally associated with mental health, we find that sexual assault, stalking, and (tentatively) sexual harassment are associated with depression and anxiety. We find that the impact of sexual orientation on anxiety is partially explained by LGBTQ+ students’ heightened exposure to sexual harassment, women’s anxiety is partially explained by experiences of sexual coercion, and law students’ depression is partially explained by experiencing sexual harassment in the form of unwanted sexual contact. Causal ordering cannot be established here, but these findings are in line with longitudinal studies that demonstrate the effects of harassment and sexual victimization on the mental health of undergraduate students (e.g., Culatta, Clay-Warner, Boyle, & Oshri, 2020; McGinley et al., 2016). While mental illness is a concern for all postbaccalaureate students, it is of particular concern for minoritized students and those who experience interpersonal violence.

Limitations and Future Research

The current study was unable to fully examine the intersections of specific races with various gender and sexual identities, such as Black lesbian women or Latino bisexual men. Given the additional, complex experiences of homophobic and racist forms of discrimination, harassment, and violence that some students face, research that moves beyond sexual harassment, as it is traditionally defined, is warranted. Also, students’ socioeconomic background, religion, citizenship status, and presence of a disability, and other aspects of their identities, could certainly affect their risk, psychological processing, and reporting of sexual harassment and assault. Future studies would benefit from larger samples, oversampling sexually and racially minoritized students, and interviews that better assess the complexity of intersecting identities and experiences. A snowball sample of students who have experienced sexual harassment might also better capture students who have changed graduate schools or left academia altogether due to the strains of their victimization or reporting experiences.

A larger random sample would be another way to improve the rigor and capacity for analysis in this study. We had a 13% response rate, as compared with 23% of graduate/professional students in the AAU’s campus climate study (Cantor et al., 2015) and 23% and 27% of recent studies of sexual assault and harassment, respectively, among graduate students (McMahon et al., 2021; Rosenthal et al., 2016). Our lower response rate may be due to our use of a lottery (that even nonparticipants could enter) instead of direct compensation (Cantor et al., 2015). We also cannot be sure whether all emailed students had access to or checked their email during the recruitment period, as only 66% of our sampling frame opened the email. A potentially more impactful recruitment strategy is exemplified in McMahon et al.’s (2021) multiple week outreach campaign in digital, print, and face-to-face formats. Advertisement and endorsed implementation through academic departments and administrators could improve participation. Our mention of sexual assault and harassment in recruitment materials (as mandated by the IRB) may also have deterred participants, and male participants in particular.

Our measure of sexual harassment also has limitations. We did not measure gender hostility—joking, behavior, or remarks that indicate women do not have equal status with men—which has been shown to be correlated with interpersonal violence (Ozaki & Otis, 2017) and a hostile environment (Schultz, 2018). The lack of this measure in our study could be contributing to our not finding significant differences in sexual harassment. Furthermore, our decision to overtly label and define sexual harassment in the survey instrument may have deterred participants from responding affirmatively to items. Listing behaviors without the “sexual harassment” label that meet the legal definition of that behavior, like our measurement of sexual assault, is a less biased approach that should be followed in future studies.

This is a nonprobability sample at one university, limiting the study’s generalizability. It is also likely that unmeasured local or departmental cultures influenced prevalence rates of harassment and sexual violence. Future research would benefit by measuring details about participants, their departments, and their victimization and reporting experiences at multiple sites. It is unclear who the perpetrators of these acts are or where assault and harassment occurred; groping by a classmate at a social event would be experienced very differently from being assaulted by one’s major professor while researching in the field (Koren, 2017). Examining interpersonal or institutional factors that facilitate victimization—and secondary victimization through unsupportive disclosure experiences and a lack of accountability—needs to be better explored to develop prevention and intervention efforts. While numerous studies have been conducted to understand undergraduate women’s reporting (and nonreporting) of sexual misconduct (e.g., Fisher, Daigle, Cullen, & Turner, 2003; Krebs et al., 2007), systematic inquiry of graduate and professional students has not, to our knowledge, been conducted.

The need for replication, multisite studies, and detailed, validated measures cannot be emphasized enough, given the understudied nature of graduate study and sexual misconduct. Future similar studies would benefit from using standardized measures of sexual assault and sexual harassment. We used the AAUW’s definition and methodology because it was developed to focus on the nature and regulation of harassment on college campuses (Hill & Silva, 2005) and the National Intimate Partner and Sexual Violence Survey because it implements both sexual violence and stalking (Black et al., 2011). However, future studies should consider utilizing a single, validated instrument that assesses diverse forms of sexual misconduct and assault.

Conclusion

The current study drew from a university that contains several legal degrees and hundreds of master’s, doctoral, and specialist programs in the humanities, social sciences, business, sciences, education, medicine, and beyond. Clearly, degree program is associated with students’ experiences: Students pursuing a doctoral degree are significantly more likely to be exposed to a hostile environment and stalking than master’s students, and nearly half of law students report at least one form of sexual harassment or assault. Because a number of characteristics were controlled for, such as gender, race, sexual orientation, age, marital status, years in the program, and international and first-generation student status—we can be at least somewhat tentatively assured that these differences are not entirely driven by selection bias into particular programs. Instead, differences by degree suggest that variation in programs’ organizational or subcultural factors may facilitate different forms of victimization. Still, preliminary evidence suggests doctoral students have lower rates of depression and anxiety than master’s students. This suggests that there is either something unique about the doctoral experience that is protective of mental health (such as peer support or mentoring) or, more pessimistically, that master’s students experiencing mental illness do not move on to doctoral programs as they may have originally planned.

Unfortunately, graduate and professional students may be more isolated from the campus community and less knowledgeable about campus resources—McMahon et al. (2021) found that graduate students have less confidence than undergraduates in seeking help after experiencing sexual violence. Health services for violence and mental health are generally underutilized on college campuses, and graduate students tend to be both financially insecure and more isolated than undergraduates (Cranford, Eisenberg, & Serras, 2009; Schneider, 1987). Thus, health professionals may want to work through graduate student organizations, such as LGBTQ+ centers or groups, Black or Hispanic Student Unions, or directly through academic departments, to target students in particular need of services. Assessment, prevention, and treatment are all necessary to improve minoritized students’ ability to complete work in safe environments, graduate, and embark on satisfying careers.

Work to assist graduate students is not just an issue of student services or human resources, but one of mentorship in academia. There is a vibrant and emerging field of literature on the experiences of LGBTQ+ academics and faculty or students of color (e.g., Biaggio, Orchard, Larson, Petrino, & Mihara, 2003; Bilimoria & Stewart, 2009; Brunsma, Embrick, & Shin, 2017; Peguero, 2018; Settles, Buchanan, & Dotson, 2019; Vaccaro, 2012; Zambrana et al., 2017). Ranging from personal essay, to literature review, to personal reflection, these works turn the inequality gaze inward toward our own disciplines and profession. Marrying this literature with research like that of Branch, Hayes-Smith, and Richards (2011), which outlines strategies for faculty to better respond to disclosure of interpersonal violence, and the development of working groups on campuses and in academic associations, is an excellent starting point in understanding how all graduate and professional students—and those most vulnerable—can be supported.

Additional work is needed to understand how faculty and others can better handle disclosures of these alarmingly common forms of misconduct and violence from their mentees and graduate or law students—and prevent them in the first place. Furthermore, training for faculty on how to identify and respond to disclosure of mental illness, a difficult and stigmatizing experience for students, could promote environments that buffer the negative psychological impact of school-related stressors and interpersonal violence. If it is a goal of graduate and professional faculty to attract, mentor, and retain talented students, attention to processes that facilitate the ejection of minoritized, exploited, or distressed students must be addressed.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Kaitlin M. Boyle

Author Biographies

Kaitlin M. Boyle is an assistant professor of sociology at Virginia Tech. She examines identity, emotion, mental health, and interpersonal violence. She recently published such work in Social Psychology Quarterly, Society and Mental Health, the Journal of Interpersonal Violence, and the Journal of School Violence.

Ashleigh E. McKinzie is an assistant professor in the sociology and anthropology department at Middle Tennessee State University. She researches long-term recovery from disasters, racial inequality, and violence against women. She is currently beginning data collection on violence against women in evangelical settings.

References

Allison

P. D.

(1999). Multiple regression: A primer. Newbury Park, CA: Pine Forge Press.

Andersen

J. P.

Hughes

T. L.

Zou

Wilsnack

S. C.

(2014). Lifetime victimization and physical health outcomes among lesbian and heterosexual women. PLoS ONE, 9(7), e101939.

Armstrong

E. A.

Gleckman-Krut

Johnson

(2018). Silence, power, and inequality: An intersectional approach to sexual violence. Annual Review of Sociology, 44(2), 1-24.

Baumer

E. P.

Lauritsen

J. L.

(2010). Reporting crime to the police, 1973–2005: A multivariate analysis of long-term trends in the National Crime Survey (NCS) and National Crime Victimization Survey (NCVS). Criminology, 48, 131-185.

Best

R. K.

Edelman

L. B.

Krieger

L. H.

Eliason

S. R.

(2011). Multiple disadvantages: An empirical test of intersectionality theory in EEO litigation. Law & Society Review, 45, 991-1025.

Biaggio

Orchard

Larson

Petrino

Mihara

(2003). Guidelines for gay/lesbian/bisexual-affirmative educational practices in graduate psychology programs. Professional Psychology: Research and Practice, 34, 548-554.

Bilimoria

Stewart

A. J.

(2009). “Don’t ask, don’t tell”: The academic climate for lesbian, gay, bisexual, and transgender faculty in science and engineering. NWSA Journal, 21, 85-103.

Black

M. C.

Basile

K. C.

Breiding

M. J.

Smith

S. G.

Walters

M. L.

Merrick

M. T.

. . . Stevens

M. R. S.

(2011). National Intimate Partner and Sexual Violence Survey. Atlanta, GA: National Center for Injury Prevention and Control, Division of Violence Prevention, Centers for Disease Control and Prevention.

Boyle

K. M.

McKinzie

A. E.

(2015). Resolving negative affect and restoring meaning: Responses to deflection produced by unwanted sexual experiences. Social Psychology Quarterly, 78, 151-172.

10.

Boyle

K. M.

Walker

L. S.

(2016). The neutralization and denial of sexual violence in college party subcultures. Deviant Behavior, 37, 1392-1410.

11.

Branch

K. A.

Hayes-Smith

Richards

T. N.

(2011). Professors’ experiences with student disclosures of sexual assault and intimate partner violence: How “helping” students can inform teaching practices. Feminist Criminology, 6, 54-75.

12.

Brown

(2015, December 21). At Harvard Law, a sex-assault case that won’t go away. The Chronicle of Higher Education. Retrieved from http://www.chronicle.com/article/At-Harvard-Law-a-Sex-Assault/234675

13.

Brunsma

D. L.

Embrick

D. G.

Shin

J. H.

(2017). Graduate students of color: Race, racism, and mentoring in the white waters of academia. Sociology of Race & Ethnicity, 3, 1-13.

14.

Bubar

(2014). Indigenous women and sexual assault: Implications for intersectionality. In Weaver

(Ed.), Social issues in contemporary Native America: reflections from Turtle Island (pp. 169-186). Burlington, VT: Ashgate Publishing.

15.

Burgess-Proctor

(2006). Intersections of race, class, gender, and crime: Future directions for feminist criminology. Feminist Criminology, 1, 27-47.

16.

Button

D. M.

Worthen

M. G.

(2014). General strain theory for LGBQ and SSB youth: The importance of intersectionality in the future of feminist criminology. Feminist Criminology, 9, 270-297.

17.

Cantor

Fisher

Chibnall

S. H.

Townsend

Lee

Thomas

. . . Westat . (2015). Report on the AAU campus climate survey on sexual assault and sexual misconduct. Washington, DC: Association of American Universities. Retrieved from https://www.aau.edu/uploadedFiles/AAU_Publications/AAU_Reports/Sexual_Assault_Campus_Survey/AAU_Campus_Climate_Survey_12_14_15.pdf

18.

Clodfelter

T. A.

Turner

M. G.

Hartman

J. L.

Kuhns

J. B.

(2010). Sexual harassment victimization during emerging adulthood: A test of routine activities theory and a general theory of crime. Crime & Delinquency, 56, 455-481.

19.

Cohen

(1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Lawrence Erlbaum.

20.

Cortina

L. M.

Swan

Fitzgerald

L. F.

Waldo

(1998). Sexual harassment and assault: Chilling the climate for women in academia. Psychology of Women Quarterly, 22, 419-441.

21.

Cranford

J. A.

Eisenberg

Serras

A. M.

(2009). Substance use behaviors, mental health problems, and use of mental health services in a probability sample of college students. Addictive Behaviors, 34, 134-145.

22.

Crenshaw

(1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43, 1241-1299.

23.

Culatta

Clay-Warner

Boyle

K. M.

Oshri

(2020). Sexual revictimization: A routine activity theory explanation. Journal of Interpersonal Violence, 35(15-16), 2800-2824.

24.

De Coster

Estes

S. B.

Mueller

C. W

. (1999). Routine activities and sexual harassment in the workplace. Work and Occupations, 26, 21-49.

25.

Dennis

A. L.

(1996). Because I am Black, because I am woman: Remedying the sexual harassment experience of Black women. Annual Survey of American Law 1996(4), 555-582.

26.

Djokic

D. S.

Lounis

(2014, April 27). This is your mind on grad school. Berkeley Science Review. Retrieved from http://berkeleysciencereview.com/article/mind-grad-school/

27.

Eisenberg

M. E.

Lust

Mathiason

M. A.

Porta

C. M.

(2017). Sexual assault, sexual orientation, and reporting among college students. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260517726414

28.

Fisher

B. S.

Cullen

F. T.

Turner

M. G.

(2000). The sexual victimization of college women: Findings from two national-level studies. NIJ Research Report (NCJ 182369). Washington, DC: U.S. Department of Justice.

29.

Fisher

B. S.

Daigle

L. E.

Cullen

F. T.

Turner

M. G.

(2003). Reporting sexual victimization to the police and others: Results from a national-level study of college women. Criminal Justice and Behavior, 30, 6-38.

30.

Gross

A. M.

Winslett

Roberts

Gohm

C. L.

(2006). An examination of sexual violence against college women. Violence Against Women, 12, 288-300.

31.

Hanson

Richards

(2017). Sexual harassment and the construction of ethnographic knowledge. Sociological Forum, 32, 587-609.

32.

Harris

H. N.

Valentiner

D. P.

(2002). World assumptions, sexual assault, depression, and fearful attitudes toward relationships. Journal of Interpersonal Violence, 17, 286-305.

33.

Hassija

C. M.

Gray

M. J.

(2013). Adaptive variants of controllability attributions among survivors of sexual assault. International Journal of Cognitive Therapy, 6, 342-357.

34.

Hernandez

T. K.

(2000). Sexual harassment and racial disparity: The mutual construction of gender and race. Journal of Gender, Race & Justice, 4, 183-244.

35.

Hill

Silva

(2005). Drawing the line: Sexual harassment on campus. American Association of University Women Educational Foundation. Retrieved from http://files.eric.ed.gov/fulltext/ED.pdf489850

36.

Johnson

P. A.

Widnall

S. E.

Benya

F. F.

(Eds.). (2018). Sexual harassment of women: Climate, culture, and consequences in academic sciences, engineering, and medicine. Washington, DC: National Academies Press.

37.

Kalof

Eby

K. K.

Matheson

J. L.

Kroska

R. J.

(2001). The influence of race and gender on student self-reports of sexual harassment by college professors. Gender & Society, 15, 282-302.

38.

Kerr

D. L.

Santurri

Peters

(2013). A comparison of lesbian, bisexual, and heterosexual college undergraduate women on selected mental health issues. Journal of American College Health, 61, 185-194.

39.

Koren

(2017, October 11). When scientists are sexually harassed in the field. The Atlantic. Retrieved from https://www.theatlantic.com/science/archive/2017/10/sexual-harassment-fieldwork-science/542559/

40.

Krebs

C. P.

Lindquist

Berzofsky

Shook-Sa

Peterson

Planty

. . . Stroop

(2016). Campus climate survey validation study final technical report. Retrieved from http://www.bjs.gov/index.cfm?ty=pbdetail&iid=5540

41.

Krebs

C. P.

Lindquist

C. H.

Warner

T. D.

Fisher

B. S.

Martin

S. L.

(2007). The Campus Sexual Assault (CSA) study: Final report. Washington, DC: National Institute of Justice, U.S. Department of Justice.

42.

Lenhart

S. A.

(2004). Clinical aspects of sexual harassment and gender discrimination: Psychological consequences and treatment interventions. New York, NY: Routledge.

43.

Lerman

L. G.

(2006). First do no harm: Law professor misconduct toward law students. Journal of Legal Education, 56, 86-105.

44.

Louros

(2016, July 18). Elephant in the room: Mental health and substance abuse in law school. American Bar Association. Retrieved from http://abaforlawstudents.com////mental-health-and-substance-abuse-in-law-school/20160718

45.

Lovibond

S. H.

Lovibond

P. F.

(1995). Manual for the Depression Anxiety Stress Scales (DASS). Sydney, New South Wales: Psychology Foundation of Australia.

46.

Mangan

(2015, July 17). Two graduate students sue UCLA over professor’s alleged sexual harassment. The Chronicle of Higher Education. Retrieved from https://www.chronicle.com/article/2-Graduate-Students-Sue-UCLA/231701

47.

McGinley

Wolff

J. M.

Rospenda

K. M.

Liu

Richman

J. A.

(2016). Risk factors and outcomes of chronic sexual harassment during the transition to college: Examination of a two-part growth mixture model. Social Science Research, 60, 297-310.

48.

McKinney

Olson

C. V.

Satterfield

(1988). Graduate students’ experiences with and responses to sexual harassment: A research note1. Journal of Interpersonal Violence, 3, 319-325.

49.

McMahon

O’Connor

Seabrook

(2021). Not just an undergraduate issue: Campus climate and sexual violence among graduate students. Journal of Interpersonal Violence, 36(7-8), NP4296-NP4314.

50.

Miranda

Jr. Meyerson

L. A.

Long

P. J.

Marx

B. P.

Simpson

S. M.

(2002). Sexual assault and alcohol use: Exploring the self-medication hypothesis. Violence and Victims, 17, 205-217.

51.

Mohipp

Senn

C. Y.

(2008). Graduate students’ perceptions of contrapower sexual harassment. Journal of Interpersonal Violence, 23, 1258-1276.

52.

Mohler-Kuo

Dowdall

G. W.

Koss

M. P.

Wechsler

(2004). Correlates of rape while intoxicated in a national sample of college women. Journal of Studies on Alcohol, 65(1), 37-45.

53.

Mustaine

E. E.

Tewksbury

(2002). Sexual assault of college women: A feminist interpretation of a routine activities analysis. Criminal Justice Review, 27, 89-123.

54.

Ozaki

Otis

M. D.

(2017). Gender equality, patriarchal cultural norms, and perpetration of intimate partner violence: Comparison of male university students in Asian and European cultural contexts. Violence Against Women, 23, 1076-1099.

55.

Peguero

A. A.

(2018). Reflections of a Latino associate professor. Hispanic Journal of Behavioral Sciences, 40, 73-80.

56.

Porter

S. R.

Whitcomb

M. E.

(2005). Non-response in student surveys: The role of demographics, engagement and personality. Research in Higher Education, 46, 127-152.

57.

Potter

(2013). Intersectional criminology: Interrogating identity and power in criminological research and theory. Critical Criminology, 21, 305-318.

58.

Resko

S. M.

(2007). Intimate partner violence against women: Exploring intersections of race, class and gender (Doctoral dissertation). The Ohio State University, Columbus.

59.

Rosenthal

M. N.

Smidt

A. M.

Freyd

J. J.

(2016). Still second class: Sexual harassment of graduate students. Psychology of Women Quarterly, 40, 364-377.

60.

Rothstein

(2007). Law students and lawyers with mental health and substance abuse problems: Protecting the public and the individual. University of Pittsburgh Law Review, 69, 531-566.

61.

Schneider

B. E.

(1987). Graduate women, sexual harassment, and university policy. The Journal of Higher Education, 58, 46-65.

62.

Schultz

(2018). Reconceptualizing sexual harassment, Again. Yale Law Journal Forum, 128. Retrieved from https://www.yalelawjournal.org/forum/reconceptualizing-sexual-harassment-again

63.

Schuman

(2014, December 4). The campus alcohol problem that nobody talks about. Slate. Retrieved from https://slate.com/human-interest/2014/12/drinking-problem-in-academia-alcohol-abuse-among-faculty-members.html

64.

Settles

I. H.

Buchanan

N. T.

Dotson

(2019). Scrutinized but not recognized: (In)visibility and hypervisibility experiences of faculty of color. Journal of Vocational Behavior, 113, 62-74.

65.

Shrier

D. K.

(1996). Sexual harassment in the workplace and academia: Psychiatric issues. Washington, DC: American Psychiatric Association.

66.

Simmons

J. P.

Nelson

L. D.

Simonsohn

(2013, January 17-19). Life after p-hacking. Meeting of the Society for Personality and Social Psychology, New Orleans, LA. Retrieved from https://ssrn.com/abstract=2205186

67.

Soper

(2006-2018). A-priori sample size calculator for multiple regression. Retrieved from https://www.danielsoper.com/statcalc/calculator.aspx?id=1

68.

Spitzberg

B. H.

Cupach

W. R.

(2007). The state of the art of stalking: Taking stock of the emerging literature. Aggression and Violent Behavior, 12, 64-86.

69.

Tang

C. S. K.

Yik

M. S.

Cheung

F. M.

Choi

P. K.

K. C.

(1996). Sexual harassment of Chinese college students. Archives of Sexual Behavior, 25, 201-215.

70.

Tarling

Morris

(2010). Reporting crime to the police. The British Journal of Criminology, 50, 474-490.

71.

Tester

(2008). An intersectional analysis of sexual harassment in housing. Gender & Society, 22, 349-366.

72.

Tjaden

Thoennes

(2000). Full report of the prevalence, incidence, and consequences of violence against women: Findings from the National Violence against Women Survey (No. NCJ 183781). Washington, DC: U.S. Department of Justice.

73.

Torrey

Ries

Spiliopoulos

(1997). What every first-year female law student should know. Columbia Journal of Gender & Law, 7, 267-311.

74.

Vaccaro

(2012). Campus microclimates for LGBT faculty, staff, and students: An exploration of the intersections of social identity and campus roles. Journal of Student Affairs Research and Practice, 49, 429-446.

75.

Van Roosmalen

E. V.

McDaniel

S. A

. (1999). Sexual harassment in academia: A hazard to women’s health. Women & Health, 28(2), 33-54.

76.

Wilson

(2012, April 18). U. of Idaho settles with family of graduate student shot dead by professor. The Chronicle of Higher Education. Retrieved from http://www.chronicle.com/article/U-of-Idaho-Settles-With/131590

77.

Wood

Hoefer

Kammer-Kerwick

Parra-Cardona

J. R.

Busch-Armendariz

(2018). Sexual harassment at institutions of higher education: Prevalence, risk, and extent. Journal of Interpersonal Violence. Advance online publication. doi:10.1177/0886260518791228

78.

Zambrana

R. E.

Dávila

B. A.

Espino

M. M.

Lapeyrouse

L. M.

Valdez

R. B.

Segura

D. A.

(2017). Mexican American faculty in research universities: Can the next generation beat the odds? Sociology of Race and Ethnicity, 3, 458-473.