A New Model for Sexual Assault Protection

Abstract

Sexual assault self-protection programs often address either broad educational goals (e.g., alcohol awareness, gender, and safety) or are restricted to the practice of violent hands-on self-protection techniques. Enrollment is almost entirely restricted to female audiences, in spite of a high risk of assault among gay men. We describe the development of Elemental, a sexual assault protection program, wherein we undertook a sociologically grounded yet multidisciplinary approach to produce a holistic and inclusive program that teaches a variety of response options, including nonviolent physical and verbal techniques. Through the use of survey data from program participants and a control group, we present results of initial longitudinal tests of the efficacy of the program. Directions for further testing and development are discussed.

Keywords

program evaluation sexual assault

Sexual assault on college campuses is a source of considerable concern and a frequent target of prevention programming efforts. These programs vary widely, with some focusing on teaching students about the contextual dangers associated with assaults (Pizone-Glover, Gidycz, and Jacobs 1998), some on changing the attitudes and behaviors of both victims and perpetrators (B. Black et al. 2000; Lonsway and Kothari 2000), others on educating students about effective ways of giving and getting sexual consent (Catharsis Productions 2014), and still others on self-defense training (Brecklin and Ullman 2005). Programs rarely, however, integrate educational programming on contextual dangers, attitudes, behaviors, and sexual consent with training in self-protection techniques (see Orchowski, Gidycz, and Raffle 2008 for an exception). Moreover, evaluations of existing programs consistently find that they have only short-term effects on students’ attitudes about assault and rarely affect rates of assault (Heppner et al. 1995; Vladutiu, Martin, and Macy 2011). Given this background, we developed a new kind of sexual assault program called Elemental.

Drawing on the expertise of sociologists, psychologists, historians, health scientists, and martial artists, we, along with an interdisciplinary team of 15 university students, crafted a program that focuses on both stranger and acquaintance assaults, combines educational and self-defense programming, provides choices in the way individuals respond to an assault, and offers an organized, holistic curriculum that is grounded in social science research and is inclusive of a variety of participants, regardless of gender, biological sex, or sexual orientation. Moreover, early indicators suggest that the program strongly affects students’ awareness of potential dangers and perceptions of sexual assault and, by doing so, lowers incidences of assault over time. In this article, we review the literature on sexual assault to illustrate gaps in existing protection programming and then describe our efforts to address those gaps through the development of Elemental. Relatedly, we discuss the program’s efficacy based on pre- and posttest data, longitudinal follow-up data, and comparison data from a nonparticipant control group.

The Prevalence and Predictors of Sexual Assault

Gender and biological sex are among the most important predictors of the risk of sexual assault. In a recent study, the Center for Disease Control (CDC) suggests that nearly 20 percent of all women experience a rape across their lifetime (defined as forcibly completed or attempted vaginal, oral, or anal penetration). Moreover, 13 percent of women experience sexual coercion (unwanted sexual penetration after being verbally, but not physically, pressured to engage in sex) and nearly 30 percent experience unwanted sexual contact such as kissing, fondling, or groping (M. C. Black et al. 2011). These figures are similar to other studies on rape and sexual aggression, many of which focus predominantly on college-aged populations. For instance, rates of rape on college campuses are commonly reported to be between 15 and 25 percent of the female student population, whereas rates of sexual coercion and unwanted sexual contact are near 11 and 27 percent, respectively (Brener et al. 1999; Fisher, Cullen, and Turner 2000; Koss, Gidycz, and Wisniewski 1987; Romeo 2004). And although less frequently studied, rates of unwanted sexual experiences for college-aged men are reported to be between 16 and 24 percent of male students (Lottes 1991; O’Sullivan, Byers, and Finkelman 1998; Struckman-Johnson 1988; Struckman-Johnson, Struckman-Johnson, and Anderson 2003; Tewksbury and Mustaine 2001). Moreover, given that individuals tend to underreport assault, it is likely that these figures underestimate the true nature of the problem for both men and women.

A growing body of research also suggests that sexual orientation is an important predictor in the risk of sexual assault (Duncan 1990; Martin et al. 2011; Menning and Holtzman 2014; Rothman and Silverman 2007). For instance, Martin et al. (2011) report a higher prevalence of sexual assault among lesbian and bisexual women, both before and during their years in college, than heterosexual women. And, in contrast to the common belief that lesbians have a lower overall risk of sexual assault relative to heterosexual women (Descamps et al. 2000), research finds that lesbians actually face the same or higher risk of assault across the life course (Balsam, Beauchaine, and Rothblum 2005; Hughes, Johnson, and Wilsnack 2001; Waldner-Haugrud and Gratch 1997). Likewise, gay and bisexual men experience assaults at higher rates than heterosexual men. Balsam et al. (2005) note that 1.6 percent of heterosexual men, 11.6 percent of gay men, and 13.2 percent of bisexual men report being raped (see also Hickson et al. 1994 and Waterman, Dawson, and Bologna 1989). Most recently, the CDC reports that rates of sexual assault among gay and bisexual men are as high as 40.2 and 47.4 percent, respectively (Walters, Chen, and Breiding 2013). These findings lead some researchers to conclude that “for men, even more so than women, a sexual minority orientation is associated with higher rates of sexual victimization” (Balsam et al. 2005, p. 484).

Moreover, researchers posit that a number of other contextual factors contribute to high rates of sexual assault, including college party culture, gender socialization, and alcohol use. For instance, studies suggest that the risk of rape for college women is related to college party culture and the degree to which men, especially those associated with fraternities, control the party environment (Armstrong, Hamilton, and Sweeny 2006; Tewksbury and Mustaine 2001). Gender socialization and gender norms promoting niceness and deference in women then interact with sexual scripts that confound seduction and acquaintance rape, thereby making it difficult for women to resist the pressure for sex or difficult for them to recognize a sexual encounter that qualifies as assault once it has occurred (Armstrong et al. 2006; Edwards et al. 2011; Ryan 2011). Moreover, the rampant alcohol abuse that occurs on college campuses, the relative insensitivity of students to the dangers posed by heavy alcohol consumption, and the well-documented link between alcohol use and rape further contribute to the rates of assault on college campuses (Abbey, McAuslan, and Ross 1998; Abbey et al. 2002; Gidycz et al. 2007; Menning 2009; Tewksbury and Mustaine 2001; Ullman and Najdowski 2010).

Existing Protection Programming for College-aged Populations

Efforts to combat the high incidences of sexual assault on college campuses are extensive. Existing programs vary in the degree to which they focus on education or self-defense, stranger or acquaintance assaults, and female or male victims. And while these approaches have their strengths, important aspects of effective sexual assault protection may be lost when programs are narrowly tailored in these ways. Below, we consider that possibility further.

Program Types: Educational Programming versus Self-defense Training

Protection programming on college campuses generally takes two forms: (1) those programs that aim to change students’ perceptions of, knowledge about, and attitudes toward sexual assault and sexual consent (see Breitenbecher and Scarce 2001; Heppner et al. 1995; Pizone-Glover et al. 1998; Vladutiu et al. 2011) and (2) those programs that teach students specific physical defenses that can be used to thwart an assault in progress (see Brecklin and Ullman 2005). Most programming on college campuses is of the first sort—it relies on lecture, discussion, and videos to inform students about the prevalence of sexual assault, the contextual features of college life that put them at risk (i.e., college party culture), and the protective strategies that can be used in sexually charged situations (B. Black et al. 2000; Breitenbecher and Gidycz 1998; Frazier, Valtinson, and Candell 1994; Gidycz, Layman et al. 2001; Gidycz, Lynn et al. 2001; Hanson and Gidycz 1993; Johnson and Russ 1989). Research consistently finds, however, that these programs are unsuccessful at producing long-term attitude changes among participants and, by extension, do not generally change behaviors or lower incidences of assault. In fact, follow-up studies with participants consistently demonstrate that within nine weeks of program completion, students’ attitudes typically rebound to pre-program levels (Breitenbecher and Gidycz 1998; Breitenbecher and Scarce 1999; 2001; Frazier et al. 1994; Gidycz, Layman et al. 2001; Gidycz, Lynn et al. 2001; Heppner et al. 1995; Lonsway and Kothari 2000). Thus, education-only programs rarely achieve their goal of producing long-term attitude change or reducing incidences of assault.

Self-defense programs, in contrast, focus less on changing attitudes toward assault and more on teaching students—typically women—how to defend themselves against an assault. Grounded in various martial arts traditions, these programs often assume that the threats women face are violent in nature and thus best answered with equally violent responses. Women, thus, learn how to effectively use their own fists, knees, fingers, and elbows as potential weapons and how to target vulnerable places on their assailant, including the eyes, nose, throat, and groin. Relatedly, women learn to yell loudly if threatened; to be assertive and direct; to avoid pleading, begging, or trying to reason with an assailant; and to fight back if at all possible (Bart and O’Brien 1981, 1985; Brecklin and Ullman 2005; Ullman 1998). Although studies suggest that women with self-defense training are more effective at thwarting an attempted rape (Bart and O’Brien 1981, 1985; Brecklin and Ullman 2005; Raleigh 2013; Ullman 1998), these programs offer very little information about the contextual dangers associated with an assault.

Taken together, then, research on existing protection programs suggests that they are only partially successful. Educational programs change attitudes and increase awareness among students, but the effects are largely temporary. Self-defense programs are associated with fewer completed rapes among female participants, but these kinds of programs are tailored to violent attacks and correspondingly violent responses and give sparse attention to other contextual risk factors associated with assault. Perhaps what is needed is a program that combines these approaches. It may be possible to more effectively impact attitudes, behaviors, and incidences of assault if students simultaneously learn to recognize and protect themselves from potential dangers not only after they escalate but also sooner in the interaction, not long after they manifest.

Assailant Types: Addressing Both Stranger and Acquaintance Assaults

One of the benefits of combining educational programming with self-defense training is the capacity to more fully address the complexities associated with the type of assailant a person may face. While educational programming often focuses predominantly on acquaintance assaults, self-defense training is typically concerned with thwarting the kinds of violent physical attacks associated with stranger assaults. To provide students with complete information about the nature of the threats they face, programming should account for both types of aggressors. But this cannot be accomplished by simply combining educational and self-defense curriculums as they are currently taught, primarily because violent protective strategies are not always well suited to the threats individuals face from acquaintances. Research suggests, for instance, that in acquaintance assaults, women do not always feel comfortable using strong physical responses (Norris, Nurius, and Dimeff 1996; Nurius 2000; Nurius et al. 2000). This is true, in part, because deeply engrained gender norms regarding politeness and meeting the needs of others, as well as fears of embarrassment or of being rejected, create psychological barriers to using self-defense tactics (Breitenbecher and Scarce 2001; Nurius 2000). Moreover, violent physical approaches may not be necessary or feel appropriate when one’s aggressor is a friend, long-term boyfriend or girlfriend, or classmate.

Therefore, efforts must be made to expressly address the degree to which sexual threats and possible responses differ depending on the type of assailant one is facing. Programming should combine educational information on both strangers and acquaintances with self-protection training that is versatile enough to provide an array of possible responses to potential sexual threats. Participants need choices in how they respond, some violent and some nonviolent, some verbal and some physical. Incorporating changes of this sort could potentially improve the long-term outcomes of program participation.

Victim Types: The Relevance of Gender, Biological Sex, and Sexual Orientation

Just as most self-defense programs do not adequately address the complexities of acquaintance assault, they also frequently restrict attendance to only women, and thus do little to offer men, especially gay and bisexual men, the opportunity to learn to protect themselves from assault. Because men are the perpetrators of most sexual violence, their exclusion from protection programs is often justified as a way to ensure they will not be able to thwart women’s self-defense techniques (Gidycz et al. 2006). However, a nontrivial number of men experience sexual assault during their lifetime, especially if they are gay or bisexual (Walters et al. 2013). Attempts to combat sexual assault, therefore, must account for and include men as victims and must acknowledge the role of sexual orientation in the experience of assault. To do otherwise means that the full scope of sexual assault will not be adequately addressed.

Developing Elemental

From all of this, then, we saw a need for a new kind of sexual assault protection program—one that combines educational and self-defense programming,offers participants choice in the way they respond to situations, accounts adequately for the contexts associated with both stranger and acquaintance assaults, and is inclusive of all individuals, regardless of sex, gender, or sexual orientation.

Designing the Educational Component of Elemental

For the educational aspects of the program, we draw on the insights of sociology, psychology, health science, and gender studies to better understand the contextual factors associated with sexual assault, including college party cultures (Armstrong et al. 2006; Menning 2009), gendered norms regarding cross-sex and same-sex interactions within dating situations (Lottes 1991; Ryan 2011; Tewksbury and Mustaine 2001; Waldner-Haugrud and Gratch 1997), the role of alcohol in sexual assault (Abbey et al. 1998; Abbey et al. 2002; Gidycz et al. 2007), and the complexities associated with asking for and giving sexual consent (Edwards et al. 2011; Struckman-Johnson et al. 2003).

This information is incorporated into a series of multimedia interactive videos that set up common sexual assault scenarios for college students, including those that could occur between strangers, long-term friends, long-term dating partners, gay men, and individuals who do not know each other well but are interested in hooking-up. By viewing these videos and dialoguing about them with the seminar instructor and with each other, Elemental participants are introduced to the contextual factors associated with assault. Following this dialogue, students engage in a variety of role-plays during which they practice verbal and physical defense skills with trained attackers (or “creepers,” as they are known colloquially within the program).

Designing the Self-defense Component of Elemental

These scenarios, thus, not only serve as the jumping-off point for important discussions during the seminar but also serve as examples of the threats that participants learn to defend against. For instance, after watching and discussing the stranger scenario, students learn a number of physical techniques that could be used if they find themselves in a situation similar to that depicted in the video. These techniques include palm-heel, knee, and elbow strikes, all of which are fairly standard self-defense skills.

For the remaining four scenarios, all of which involve acquaintances, purely physical responses are not sufficient. Given the research suggesting some women are uncomfortable using strong physical responses when they are dealing with acquaintances, we include both verbal and nonviolent physical responses to the depicted threats. Moreover, the verbal techniques go beyond teaching students to loudly yell “No!” or “Leave me alone!” Instead, we rely heavily on research in social psychology and incorporate verbal strategies that promote shutting down potentially problematic situations early. For instance, work within the psychology of influence, including research on the principles of consistency and reciprocity (Cialdini 2009; Festinger 1962; Greenwald et al. 1987; Gouldner 1960; Regan 1971), and research in psycholinguistics on the importance of particular language cues for increasing compliance with one’s wishes (Howard, Gengler, and Jain 1995; Kimps and Davidse 2008; Langer, Blank, and Chanowitz 1978) are incorporated into Elemental’s curriculum. These verbal techniques, in conjunction with the physical responses that vary in level and intensity of violence, ensure that the program offers participants a number of options for dealing with potentially threatening sexual situations.

Designing an Array of Verbal, Physical, Violent, and Nonviolent Response Options

One of the benefits of designing a program that offers participants choice in how they react to sexual threats is that it allows them to fashion responses based on their own feelings, personalities, and assessments of the situation. One of the challenges of designing such a program, however, is ensuring that choice is built into the program in a number of fairly nuanced ways. For instance, it is not sufficient to teach students only one verbal and one physical response to a given threat, for doing so may continue to limit their ability to react in ways that are consistent with their own feelings and personalities. We contend that effective self-protection must be flexible enough to offer choices within the verbal and physical response options.

To accomplish this, we use To-Shin Do as the foundation for all the self-defense strategies, both verbal and physical, violent and nonviolent, that we address in Elemental. To-Shin Do is characterized by a focus on achieving one’s goals, rather than dominating one’s opponent (Hayes 2012). This perspective is useful in unwanted sexual encounters, especially those between acquaintances, because it allows a person to avoid sex while still preserving the relationship with the other person, if they so desire. To-Shin Do use a problem-solving paradigm that is grounded in four elements: earth, water, fire, and wind. This elemental approach to problem-solving allows a person to react to a situation based on how he or she is feeling—confident (earth), surprised (water), connected (fire), or withdrawn (wind). Thus, for each of the sexual threats portrayed in the videos, we teach both verbal and physical responses consistent with each of the four elements of To-Shin Do. These elemental responses are not only the namesake for the program but they are a cornerstone of its uniqueness precisely because they offer participants a number of choices in how they respond to sexual threats. Ultimately, this gives students the ability to fashion responses that make sense to them and are consistent with how they are feeling if they face unwanted sexual pressures.

Aims of the Study

Program design and refinement have taken more than three years so far, but we are now in a position to examine program efficacy using both longitudinal data from participants as well as cross-sectional comparison data from a nonparticipant control group. These data allow us to assess the degree to which Elemental produces long-term change in students’ attitudes and behaviors regarding sexual assault as well as the degree to which the program lowers the risk of assault over time. Although all findings are considered preliminary, and data collection is ongoing, findings thus far are encouraging.

Method

Program Group

We recruited students to participate in Elemental via a series of emails and other announcements that were circulated to all incoming freshmen, regardless of sex or sexual orientation, at two post-secondary institutions in the Midwest (one large university and one small liberal arts college) at the beginning of the fall semester of 2012.¹ At the beginning of each of seven sessions of the program, which were conducted during the first month of the semester at no cost to students, we invited all those attending to participate in a longitudinal research study, consisting of a pretest, a posttest, and a series of follow-up surveys. Ninety-five of the 110 participants (86 percent) signed up for the research study. We combined this data with that from a small pilot sample of individuals (12) who took the program in the spring of 2012 and used an identical instrument. Thus, our program assessment is based on data from 107 participants, all of whom completed the pretest and posttest during their participation in Elemental (administered at the beginning and end of the seminar, respectively). These individuals were then invited, via email, to participate in a six-month online follow-up survey during the fall of 2012 (pilot group) or the spring of 2013. We incentivized participation with a drawing for one of three $50 prizes. Sixty-six students (62 percent) participated in the longitudinal follow-up survey.

Control Group

During the same period of time that the follow-up survey was available, we asked students in an introductory sociology course to complete a voluntary, one-time survey that was administered during class. From this, we garnered a control-group sample of 150 students. Comparisons for the control-group and program-group samples reveal no significant differences on demographic variables such as age, race, and year in school. Except in those instances where questions are program-specific, items on the control-group survey are identical to those used in the instruments administered to the program group. All procedures for both the program group and the control group were conducted in accordance with an Institutional Review Board-approved human subjects protocol.

Measures

See Table 1 for basic descriptive statistics for variables used in pooled-sample analyses.

Table 1.

Descriptive Statistics (N = 192 [48 participants, 144 control]).

	Participants		Control
Variable	$\bar{x}$	SD	$\bar{x}$	SD
Sexual Assault Self-protection Scale	4.146	0.542	3.693	0.643
Sexual assault	0.083	0.279	0.063	0.243
Male	0.021	0.144	0.340	0.475
Academic year	1.646	1.041	1.569	0.781
Previous assault	0.500	0.502	0.229	0.422
Previous martial arts instruction	0.167	0.377	0.146	0.354

Sexual Assault Self-protection Scale

The pretest, posttest, and control surveys ask participants a number of attitudinal questions about sexual assault, sexual consent, personal confidence, contextual awareness, and boundary setting that we believe are important for effective self-protection from sexual assault. Preliminary analyses suggest that a subset of 17 of these items factor together well (Cronbach’s α = 0.84) to create what we call the Sexual Assault Self-protection Scale. Items on this scale measure levels of agreement or disagreement (Likert-type scale: 1 = strongly disagree, 7 = strongly agree) with the statements listed in Table 2.

Table 2.

Sexual Assault Self-protection Scale items.

Item No.	Text
1	I am comfortable saying, “no” when I do not want something.
2	I stand up for myself.
3	I protect my personal space effectively.
4	I am comfortable setting boundaries.
5	If someone tried to hurt me, I would freeze.^a
6	It is important to make others happy even if it makes me unhappy.^a
7	Strangers are always more dangerous than people I know.^a
8	I am assertive.
9	I am good at keeping people from pressuring or manipulating me into doing things that I do not want to do.
10	I do things because I feel obligated to do them, even if it is not good for me personally.^a
11	Alcohol can be dangerous in social situations.
12	When it comes to sexual issues, I understand what consent is.
13	When it comes to sex, I am confident in my ability to communicate what I want and do not want.
14	I am aware of what happens around me.
15	I am good at watching out for myself.
16	I know what to look for to tell if a situation puts me at risk of sexual assault.
17	I know what to do when a situation puts me at risk of sexual assault.

Items reverse coded (e.g., 5 becomes −5) to maintain consistency of scale direction.

The resulting scale represents the mean scores of all items for a given respondent. For the sample of Elemental program participants (at posttest) and the control sample, the Sexual Assault Self-protection Scale scores range from 0.882 to 5.118 ( $\bar{x}$ = 3.843, SD = 0.645).

Elemental program participant

A binary variable measuring program participation (1 = program participant, 0 = control group) is included in models measuring the effects of program participation.

Sexual assault

All students were asked if they had experienced unwanted sexual contact or activity that was awkward, dangerous, or sexually charged (1 = yes, 0 = no) since starting college (control group) or since taking Elemental (program group at follow-up).

Controls

In all inferential models, control measures are included for sex (1 = male, 0 = female), academic year (1 = freshman, 2 = sophomore, 3 = junior, 4 = senior), sexual assault prior to college (1 = yes, 0 = no), and previous martial arts instruction (1 = yes, 0 = no). Supplementary analysis using a binary logistic regression model of program participation on controls indicates that program participants are less likely to be male (OR = 0.048; p < 0.01) and more likely to have experienced a sexual assault prior to college (OR = 2.584; p < 0.01) than individuals in the control group. (Results available upon request.) This latter finding is particularly noteworthy, as research suggests that a prior assault tends to raise the risk of later assault (Breitenbecher and Scarce 2001). As such, the program group represents those at particularly high risk, both due to their biological sex and their assault histories. This is an issue we address further in the conclusion of the article.

Analytic Strategy

Given that most existing protection programs aim to reduce the risk of sexual assault by changing students’ attitudes about assault, consent, and risky behavior, we first examine whether or not participation in Elemental influences students’ attitudes between the pretest and posttest measures and also relative to a nonparticipant control group. In conjunction, we examine whether any influence that Elemental might have on students’ attitudes is, in turn, associated with a lower risk of assault six months after completing the program. Said another way, we test the following set of relationships:

Figure 1.

Expected Relationship Links.

We examine these relationships by using a combination of statistical models and descriptive analyses. More specifically, we use a series of paired-sample t tests to test whether program participation affects scores on an attitude scale of our own design (the Sexual Assault Self-protection Scale, described in depth below), and we use ordinary least squares (OLS) regression to test whether program participants at posttest had higher scale scores than those of nonparticipants. To validate the scale, we then use binary logistic regression to test whether scale scores are predictive of later sexual assault.

Results

The pretest/posttest results indicate clear, statistically significant changes (p < 0.05) in program participants’ scores on all of the individual items on the Sexual Assault Self-protection Scale, on the scale overall, and in the expected direction of change (see Table 3). Overall, participation in Elemental does influence students’ assault-related knowledge and attitudes, such that their scores on the Sexual Assault Self-protection Scale increase by an average of over 1.5 points.

Table 3.

Change in Sexual Assault Self-protection Scale and Component Scores, Elemental Program Pretest to Posttest (N = 107).

Item	${\bar{x}}_{pretest}$	${\bar{x}}_{posttest}$
1	5.678	6.383***
2	5.579	6.182***
3	5.402	6.136***
4	5.650	6.201***
5^a	3.750	2.877***
6^a	3.650	2.486***
7^a	3.626	3.009***
8	4.724	5.425***
9	5.140	5.804***
10^a	3.542	2.645***
11	6.304	6.621***
12	6.327	6.748***
13	5.607	6.290***
14	5.505	5.963***
15	5.556	6.093***
16	4.650	6.215***
17	4.061	6.336***
Scale	2.534	4.082***

Note. Paired sample t tests of significant difference (two-tailed) from pretest to posttest.

Item enters negatively into scale.

***

p < .001.

Moreover, a comparison of participants’ posttest scores with those taken from a control sample of nonparticipants indicates that Elemental participants have significantly higher scores than those of the student body overall, net of the effects of sex, prior victimization, or other martial arts instruction (see Table 4). Taken together, these analyses indicate that program participation has a strong effect on Sexual Assault Self-protection Scale scores. Not only do program participants see gains in the attitudes and beliefs believed to be related to sexual assault risk, but their posttest scores are higher than the Sexual Assault Self-protection Scale scores of their nonparticipant peers.

Table 4.

Effects of Elemental Program Participation and Background Variables on Sexual Assault Self-protection Scale Scores (N = 192).

Variables	Sexual Assault Self-protection Scale score
Male	0.259* (0.106)
Academic year	0.0197 (0.0514)
Previous assault	−0.171† (0.0999)
Previous martial arts instruction	0.223† (0.122)
Elemental program participant	0.577** (0.108)
Intercept	3.580** (0.106)
Observations	192
R ²	0.159

Note. Unstandardized OLS regression results with standard errors in parentheses. OLS = ordinary least squares.

†

p < .1. *p < .05. **p < .01.

The question, of course, then becomes whether or not these attitude changes are associated with a decreased risk of sexual assault. In a pooled sample of program participants and nonparticipants taken approximately six months after the program, higher scores on the Sexual Assault Self-protection Scale (with program participants’ scores taken at the posttest) are associated with a lower risk of unwanted sexual contact since taking the program (or since beginning college, for nonparticipants). This association persists net of the effects of sex, year in college, pre-college victimization, or previous martial arts experience (see Table 5). The odds ratio indicates that for an increase of one point on the Sexual Assault Self-protection Scale, the odds of sexual assault are decreased by a factor of 0.34, holding all other variables constant. Or, equivalently, for each additional point increase on the Sexual Assault Self-protection Scale, the odds of sexual assault are decreased by 66 percent. Thus, to the extent that participation in Elemental significantly influences participants’ scores on the Sexual Assault Self-protection Scale, Elemental can be said to help reduce one’s risk of assault up to six months post-seminar.

Table 5.

Relationship of Sexual Assault Self-protection Scale Scores and Background Variables to Odds of Sexual Assault since Beginning College/Taking Elemental (N = 192).

Variables	Sexual assault
Male	0.806 (0.686)
Academic year	1.473 (0.476)
Previous assault	2.529 (1.623)
Previous Martial Arts instruction	0.487 (0.537)
Self-protection Scale	0.342* (0.144)
Intercept	1.398 (2.186)
Observations	192

Note. Odds ratios from binary logistic regression results with standard errors in parentheses.

p < .05.

It appears, then, that Elemental not only influences participants’ knowledge and attitudes about sexual assault and related issues, but by virtue of that attitudinal influence, the program also lowers risk of assault for at least six months post-seminar. This is an important finding because few education-based programs have any kind of long-term influence on sexual assault attitudes or incidences.

Discussion and Conclusion

Elemental offers a unique approach to sexual assault self-protection because it is inclusive of all individuals, regardless of gender, sex, or sexual orientation, and because it combines educational programming on assault, consent, party culture, party safety, and so forth with physical and verbal self-defense training. During the six-hour seminar, students learn how to recognize sexual threats early, communicate with partners about sex, and use a variety of self-protection techniques that vary in intensity and level of violence.

Data on the program’s effectiveness, while preliminary, are encouraging. The pretest, posttest, and longitudinal follow-up data suggest that the program is effective at affecting participants’ sexual assault-related attitudes and behaviors and at lowering participants’ risk of sexual assault. More specifically, participation in the program appears to affect students’ perceptions of safety and heightens their awareness of issues associated with the risk of assault. Importantly, these effects persist for at least six months following the completion of the program. This is noteworthy because tests of existing programs consistently find that changes to students’ attitudes and behaviors are largely temporary and rebound to pre-program levels within nine weeks of program completion (Breitenbecher and Gidycz 1998; Breitenbecher and Scarce 1999, 2001; Frazier et al. 1994; Gidycz, Layman et al. 2001; Gidycz, Lynn et al. 2001; Heppner et al. 1995; Lonsway and Kothari 2000).

Moreover, because the effects of existing programs are generally temporary, they typically do not affect rates of assault. Among Elemental participants, however, increased awareness of party safety, the dangers of alcohol use, the complexities of giving and getting sexual consent, and so forth are associated with a lower risk of assault for at least six months after program completion. While our analyses do not allow us to parse out the potential individual-level or interactional effects of Elemental’s educational components and self-defense sequences, we believe it may be precisely this combination of curriculum characteristics that has produced the long-term effects we see in our data. In support of this assertion, we point to programming conducted at the Laboratory for the Study of Sexual Assault Prevention at Ohio University. Christine Gidycz and her colleagues have been assessing the effectiveness of protection programs for more than a decade. They initially began with education-only programs, but repeatedly found few, if any, long-term effects of these programs on attitudes or incidences of assault (Breitenbecher and Gidycz 1998; Hanson and Gidycz 1993; Gidycz, Layman et al. 2001; Gidycz, Lynn et al. 2001; Pizone-Glover et al. 1998). More recently, however, they have tested a program that couples educational information, role-playing exercises, and student discussion with a two-hour self-defense seminar that focuses on “assertive” protection measures. Tests of this program reveal that attitude and behavior changes among participants persisted over time (follow-ups ranged from two to six months), although incidences of assault were not affected (Gidycz et al. 2006; Orchowski et al. 2008). Gidycz and her colleagues did, however, find a difference in the severity of assaults experienced by women in the program group and the control group, with women in the control group being more likely to experience a rape while program participants were more likely to experience a more moderate form of assault. Importantly, women in the program group were also more likely to use self-protective strategies. This suggests that program participants may have been more adept at preventing the escalation of threatening situations. Gidycz’s program, then, does appear to be having a nontrivial impact on the way women experience an assault.

When considered in tandem with Gidycz’s work, the findings presented here suggest not only that Elemental, specifically, has potential but also that protection programming, in general, should consider including both educational and self-defense training components conjointly within curriculums. It may be that program effectiveness hinges on the way the educational and self-defense components mutually reinforce one another. If that is true, then it will be through this kind of combination of tactics that optimal program efficacy will be achieved.

While it is too soon to make strong policy suggestions, these findings do point to important potentials within sexual assault protection programming that need to be explored further. Moreover, the time for such exploration is now. Concern about sexual assault on college campuses is uniformly high and President Obama recently created a task force to look into the issue with the aim of developing “a coordinated Federal response to campus rape and sexual assault” (Obama 2014). More specifically, he charged the task force with, among other things, “providing institutions with evidence-based best and promising practices for preventing and responding to rape and sexual assault” (Obama 2014). Elemental is one such promising program, and it appears that combining educational and self-defense training is a promising practice as well. Universities, then, may need to reconsider their tendency toward reliance on predominantly education-only programming.

Universities should also consider adopting programs that are inclusive of a variety of individuals. Although sexual assault rates are highest among female college students, male students face threats as well. Any coordinated effort to address sexual assault, then, must account for the safety of male students. Likewise, such efforts should acknowledge and account for the role of sexual orientation. For men, especially, biological sex and sexual orientation interact in ways that significantly increase the risk of assault. If programming does not fully address this kind of complexity, it will fall short of its goals by continuing to diminish the threats men face and continuing to exclude them from protection efforts. While we recognize and appreciate that the inclusion of men in programs does introduce potential issues, including female discomfort with mixed-sex classes and the potential for male perpetrators to learn how to thwart self-protective strategies, these issues do not negate the fundamental fact that male students can be victimized, and it is inappropriate to deny them self-protection training.

Programming efforts should also further explore the potential benefits associated with offering students a broad array of response options. Although some existing programs do teach a number of different self-defense tactics, Elemental appears to be unique in the array of responses offered to students. For instance, Gidycz’s program focuses on “assertive” self-defense training, defined as “forceful physical resistance, nonforceful physical resistance, and forceful verbal resistance” (Gidycz et al. 2006:175). While this approach does give participants choices in how they respond to threatening situations, the options are still grounded largely in confrontational, forceful tactics that may not feel appropriate for all individuals in all circumstances. The findings outlined here suggest that a broader array of protective strategies may be warranted.

These policy recommendations are, of course, preliminary. Scholars should continue to test the efficacy of combined approaches, the benefits of inclusive programs, and the utility of offering an array of self-defense response options. We are currently doing so with Elemental. While the results presented here are based on six-month follow-ups, we are continuing to collect data on our program participants and will do so for up to four years post-seminar completion. We also now have initial and longitudinal data on two separate cohorts of participants. In future studies, therefore, we will be able to pool cohort data, test for longer term effects, and compare effects across cohorts that may be associated with further refinements to the program.

Moreover, we continue to refine the research protocol for the program in an effort to make our program group and control group more comparable. For instance, we recently instituted a longitudinal data collection effort with control-group students, rather than the cross-sectional design used here. This will enable us to ensure maximum comparability between the two groups of students and allow us to ensure that the attitude changes we see in our participants relative to the control group are not merely a test–retest artifact. Relatedly, we are seeking a more robust test of the links between program participation, the Sexual Assault Self-protection Scale, and risk of assault. Although the scale scores are at least temporarily affected by program participation and the scale does appear to be linked strongly with assault (and, in the model presented here, is a better predictor of assault than sex or previous assault experience), the tests could be stronger. For example, as the sample size increases, it will become possible to use mediation analyses and related techniques to more fully explore the links among the factors investigated here (Fritz and MacKinnon 2007). Augmenting these tests with the new longitudinal data from the control group and additional program group, follow-up data will allow us to track any potential effects of campus socialization and ensure that the change observed from pretest to posttest in the program group is not ephemeral.

We are also exploring ways to alter program recruitment efforts. Currently, the program is promoted to all incoming Freshmen, but students ultimately self-select to participate. This self-selection process is largely responsible for the disproportionate representation of assault survivors in the program group. This fact may raise concerns that the program group and control group are not comparable to one another and thus raise the question of bias in the sample. However, when one considers that re-victimization rates are generally higher among assault survivors (Gidycz et al. 1993), the effects that we see with our program participants are especially noteworthy. In short, given that the indicators that we have at this time suggest that program-group participants are at a higher risk of assault than their peers as they enter the program, we have strong reason to believe that any biases introduced by nonrandomness would count against the program’s effectiveness in the statistical analyses and that our findings are thus conservative. We, thus, cautiously suggest that Elemental may effectively lower risk among those whose backgrounds indicate greater vulnerability, as well as for those without prior victimization. As our sample sizes increase and our recruiting mechanisms change (e.g., incorporating the seminar into regular residence hall programming), we anticipate that the corresponding gain in statistical power and change in participants’ demographics will enable us to show more definitive links between survivors’ (and other subpopulations’) experiences in the program and their later outcomes using more advanced program evaluation techniques, including those that fully account for potential self-selection effects.

In addition, through the collection of qualitative data, we hope to expand on our understanding of the processes of sexual assault self-protection, for example, when do people choose to act, why and how do people choose or eliminate specific strategies in their self-defense, how are the chosen strategies associated with concepts of gender and self-empowerment, and by what criteria do people judge a strategy to be successful? It is, after all, through the respondents’ own judgments of success that we can come to claim that Elemental succeeds in meeting its goals of personal empowerment and justice. We hope that our program, through an odyssey of continued testing and refinement, will serve as an example of how the creative application of sociological investigation and knowledge can produce programming that is both well adapted to the social structures within which it must operate and a powerful agent in the resolution of social problems.

Footnotes

Acknowledgements

The authors wish to thank Ball State University and the Virginia Ball Center for Creative Inquiry for their generous support of this project. We also thank the many students who participated in this project, as well as Morgan Roddy, Elaina Johns-Wolfe, and Zachary Palmer for valuable research assistance.

Authors’ Note

An earlier version of this article was presented at the 2013 meetings of the Association for Applied and Clinical Sociology in Portland, Oregon.

Declaration of Conflicting Interests

Authors are co-owners of Elemental Sexual Assault Protection, LLC, a newly established company that administers sexual assault self-protection seminars and certifies others to do the same. Elemental, LLC, has not sponsored this research and was incorporated after the collection of the data presented here.

Funding

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

Notes

Author Biographies

Mellisa Holtzman is an associate professor of sociology at Ball State University. While most of her research has focused on the sociology of family and family law, recent work has extended to her emerging interests in sex, gender, and sexual assault prevention. Her publications include articles in Journal of Interpersonal Violence, Perspectives on Sexual and Reproductive Health, Law and Social Inquiry, Family Court Review, and Family Relations.

Chadwick Menning is an associate professor of sociology at Ball State University. His research interests are centered on young adult safety and well-being. He has published works in Journal of Interpersonal Violence, Journal of Health and Social Behavior, and Perspectives on Sexual and Reproductive Health.

References

Abbey

McAuslan

Ross

1998. “Sexual Assault Perpetration by College Men: The Role of Alcohol, Misperception of Sexual Intent, And Sexual Beliefs and Experiences.” Journal of Social & Clinical Psychology 17:167–95.

Abbey

Zawacki

Buck

Testa

Parks

Norris

Martin

Livingston

McAuslan

Clinton

A. M.

Kennedy

George

Davis

Martell

2002. “How Does Alcohol Contribute to Sexual Assault? Explanations from Laboratory and Survey Data.” Alcoholism: Clinical & Experimental Research 26:575–81.

Armstrong

Hamilton

Sweeny

2006. “Sexual Assault on Campus: A Multilevel Integrative Approach to Party Rape.” Social Problems 53:483–99.

Balsam

Beauchaine

Rothblum

2005. “Victimization Over the Life Span: A Comparison of Lesbian, Gay, Bisexual, and Heterosexual Siblings.” Journal of Consulting and Clinical Psychology 73:477–87.

Bart

O’Brien

1981. “A Study of Women Who Both Were Raped and Avoided Rape.” Journal of Social Issues 37:123–37.

Bart

O’Brien

1985. Stopping Rape: Successful Survival Strategies. New York: Pergamon.

Black

Weisz

Coats

Patterson

2000. “Evaluating a Psychoeducational Sexual Assault Prevention Program Incorporating Theatrical Presentation, Peer Education, and Social Work.” Research on Social Work Practice 10:589–606.

Black

M. C.

Basile

K. C.

Breiding

M. J.

Smith

S. G.

Walters

M. L.

Merrick

M. T.

Chen

Stevens

M. R.

2011. The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 Summary Report. Atlanta: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Brecklin

Ullman

2005. “Self-Defense or Assertiveness Training and Women’s Responses to Sexual Attacks.” Journal of Interpersonal Violence 20:738–62.

10.

Breitenbecher

Gidycz

1998. “An Empirical Evaluation of a Program Designed to Reduce the Risk of Multiple Sexual Victimization.” Journal of Interpersonal Violence 13:472–88.

11.

Breitenbecher

Scarce

1999. “A Longitudinal Evaluation of the Effectiveness of a Sexual Assault Education Program.” Journal of Interpersonal Violence 14:459–78.

12.

Breitenbecher

Scarce

2001. “An Evaluation of the Effectiveness of a Sexual Assault Education Program Focusing on Psychological Barriers to Resistance.” Journal of Interpersonal Violence 16:387–407.

13.

Brener

McMahon

Warren

Douglas

1999. “Forced Sexual Intercourse and Associated Health-Risk Behaviors among Female College Students in the United States.” Journal of Consulting and Clinical Psychology 67:252–59.

14.

Catharsis Productions. 2014. “Sex Signals.” Retrieved January 28, 2014 (http://www.catharsisproductions.com/sexsignals.php).

15.

Cialdini

2009. Influence: Science and Practice. Boston: Pearson.

16.

Descamps

Rothblum

Bradford

Ryan

2000. “Mental Health Impact of Child Sexual Abuse, Rape, Intimate Partner Violence, and Hate Crimes in the National Lesbian Health Care Survey.” Journal of Gay & Lesbian Social Services 11:27–55.

17.

Duncan

1990. “Prevalence of Sexual Assault Victimization among Heterosexual and Gay/Lesbian University Students.” Psychological Reports 66:65–66.

18.

Edwards

Turchik

Dardis

Reynolds

Gidycz

2011. “Rape Myths: History, Individual and Institutional-level Presence, and Implications for Change.” Sex Roles 65:761–73.

19.

Festinger

1962. A Theory of Cognitive Dissonance. Stanford: Stanford University Press.

20.

Fisher

Cullen

Turner

2000. The Sexual Victimization of College Women. Washington, DC: National Institute of Justice and the Bureau of Justice Statistics.

21.

Frazier

Valtinson

Candell

1994. “Evaluation of a Coeducational Interactive Rape Prevention Program.” Journal of Counseling & Development 73:153–58.

22.

Fritz

M. S.

MacKinnon

D. P.

2007. “Required Sample Size to Detect the Mediated Effect.” Psychological Science 18:233–39.

23.

Gidycz

Layman

Rich

Crothers

Matorin

Jacobs

2001. “An Evaluation of an Acquaintance Rape Prevention Program: Impact on Attitudes, Sexual Aggression, and Sexual Victimization.” Journal of Interpersonal Violence 16:1120–38.

24.

Gidycz

Loh

Lobo

Rich

Lynn

Pashdag

2007. “Reciprocal Relationships among Alcohol Use, Risk Perception, and Sexual Victimization: A Prospective Analysis.” Journal of American College Health 56:5–14.

25.

Gidycz

Lynn

Rich

Marioni

Loh

Blackwell

Stafford

Fite

2001. “The Evaluation of a Sexual Assault Risk Reduction Program: A Multisite Investigation.” Journal of Consulting and Clinical Psychology 69:1073–78.

26.

Gidycz

Nelson Coble

Latham

Layman

M. J.

1993. “Sexual Assault Experience in Adulthood and Prior Victimization Experiences.” Psychology of Women Quarterly 17:151–68.

27.

Gidycz

Rich

Orchowski

King

Miller

2006. “The Evaluation of a Sexual Assault Self-Defense and Risk Reduction Program for College Women: A Prospective Study.” Psychology of Women Quarterly 30:173–86.

28.

Gouldner

1960. “The Norm of Reciprocity: A Preliminary Statement.” American Sociological Review 25:161–78.

29.

Greenwald

Carnot

Beach

Young

1987. “Increasing Voting Behavior by Asking People if They Expect to Vote.” Journal of Applied Psychology 72:315–18.

30.

Hanson

Gidycz

1993. “Evaluation of a Sexual Assault Prevention Program.” Journal of Consulting and Clinical Psychology 61:1046–52.

31.

Hayes

S. K.

2012. The Ninja Defense: A Modern Master’s Approach to Universal Dangers. Rutland, VA: Tuttle Publishing.

32.

Heppner

Humphrey

Hillenbrand-Gunn

DeBord

1995. “The Differential Effects of Rape Prevention Programming on Attitudes, Behavior, and Knowledge.” Journal of Counseling Psychology 42:508–18.

33.

Hickson

Davies

Hunt

Weatherburn

McManus

Coxon

1994. “Gay Men as Victims of Nonconsensual Sex.” Archives of Sexual Behavior 23:281–94.

34.

Howard

Gengler

Jain

1995. “What’s in a Name? Complimentary Means of Persuasion.” Journal of Consumer Research 22:200–11.

35.

Hughes

Johnson

Wilsnack

2001. “Sexual Assault and Alcohol Abuse: A Comparison of Lesbians and Heterosexual Women.” Journal of Substance Abuse 13:515–32.

36.

Johnson

Russ

1989. “Effects of Salience of Consciousness-raising Information on Perceptions of Acquaintance versus Stranger Rape.” Journal of Applied Social Psychology 19:1182–97.

37.

Kimps

Davidse

2008. “Illocutionary Force and Conduciveness in Imperative Constant Polarity Tag Questions: A Typology.” Text & Talk 28:699–722.

38.

Koss

Gidycz

Wisniewski

1987. “The Scope of Rape: Incidence and Prevalence of Sexual Aggression and Victimization in a National Sample of Higher Education Students.” Journal of Consulting and Clinical Psychology 55:162–70.

39.

Langer

Blank

Chanowitz

1978. “The Mindlessness of Ostensibly Thoughtful Action: The Role of ‘Placebic’ Information in Interpersonal Interaction.” Journal of Personality and Social Psychology 36:635–42.

40.

Lonsway

Kothari

2000. “First Year Campus Acquaintance Rape Education: Evaluating the Impact of a Mandatory Intervention.” Psychology of Women Quarterly 24:220–32.

41.

Lottes

1991. “The Relationship between Nontraditional Gender Roles and Sexual Coercion.” Journal of Psychology and Human Sexuality 4:89–109.

42.

Martin

Fisher

Warner

Krebs

Lindquist

2011. “Women’s Sexual Orientation and Their Experiences of Sexual Assault Before and During University.” Women’s Health Issues 21:199–205.

43.

Menning

C. L.

2009. “Unsafe in Any House?: Attendees’ Perceptions of Microlevel Environmental Traits and Personal Safety at Fraternity and Nonfraternity Parties.” Journal of Interpersonal Violence 24:1714–34.

44.

Menning

C. L.

Holtzman

2014. “Processes and Patterns in Gay, Lesbian, and Bisexual Sexual Assault: A Multimethodological Assessment.” Journal of Interpersonal Violence 29:1071–93.

45.

Norris

Nurius

Dimeff

1996. “Through Her Eyes: Factors Affecting Women’s Perceptions of and Resistance to Acquaintance Sexual Aggression Threat.” Psychology of Women Quarterly 20:123–45.

46.

Nurius

2000. “Risk Perception for Acquaintance Sexual Aggression: A Social-cognitive Perspective.” Aggression and Violent Behavior 5:63–78.

47.

Nurius

Norris

Young

Graham

Gaylord

2000. “Interpreting and Defensively Responding to Threat: Examining Appraisals and Coping with Acquaintance Sexual Aggression.” Violence and Victims 15:187–208.

48.

Obama

2014. “Establishing a White House Task Force to Protect Students from Sexual Assault.” Retrieved May 20, 2014 (http://www.whitehouse.gov/the-press-office/2014/01/22/memorandum-establishing-white-house-task-force-protect-students-sexual-a).

49.

Orchowski

Gidycz

Raffle

2008. “Evaluation of a Sexual Assault Risk Reduction and Self-defense Program: A Prospective Analysis of a Revised Protocol.” Psychology of Women Quarterly 32:204–18.

50.

O’Sullivan

Byers

Finkelman

1998. “A Comparison of Male and Female College Students’ Experiences of Sexual Coercion.” Psychology of Women Quarterly 22:177–95.

51.

Pizone-Glover

Gidycz

Jacobs

1998. “An Acquaintance Rape Prevention Program: Effects on Attitudes toward Women, Rape-related Attitudes, and Perceptions of Rape Scenarios.” Psychology of Women Quarterly 22:605–21.

52.

Raleigh

(2013). “Are Women Safer When They Learn Self-defense? Cascade.” Retrieved May 20, 2014 (http://cascade.uoregon.edu/spring2013/social-sciences/are-women-safer-when-they-learn-self-defense/).

53.

Regan

1971. “Effects of a Favor and Liking on Compliance.” Journal of Experimental Social Psychology 7:627–39.

54.

Romeo

2004. “Acquaintance Rape on College and University Campuses.” College Student Journal 38:61–65.

55.

Rothman

Silverman

2007. “The Effect of a College Sexual Assault Prevention Program on First Year Students’ Victimization Rates.” Journal of American College Health 55:283–90.

56.

Ryan

2011. “The Relationship between Rape Myths and Sexual Scripts: The Social Construction of Rape.” Sex Roles 65:774–82.

57.

Struckman-Johnson

1988. “Forced Sex on Dates: It Happens to Men, Too.” The Journal of Sex Research 24:234–41.

58.

Struckman-Johnson

Anderson

2003. “Tactics of Sexual Coercion: When Men and Women Won’t Take No for an Answer.” The Journal of Sex Research 40:76–86.

59.

Tewksbury

Mustaine

2001. “Lifestyle Factors Associated with the Sexual Assault of Men: A Routine Activity Theory Analysis.” The Journal of Men’s Studies 9:153–82.

60.

Ullman

1998. “Does Offender Violence Escalate When Rape Victims Fight Back?” Journal of Interpersonal Violence 13:179–92.

61.

Ullman

Najdowski

2010. “Understanding Alcohol-related Sexual Assaults: Characteristics and Consequences.” Violence and Victims 25:29–44.

62.

Vladutiu

Martin

Macy

2011. “College-or University-based Sexual Assault Prevention Programs: A Review of Program Outcomes, Characteristics, and Recommendations.” Trauma, Violence, & Abuse 12:67–86.

63.

Waldner-Haugrud

1997. “Sexual Coercion In Gay/Lesbian Relationships: Descriptives and Gender Differences.” Victims and Victimology 12:87–98.

64.

Walters

Chen

Breiding

2013. The National Intimate Partner and Sexual Violence Survey (Nisvs): 2010 Findings on Victimization by Sexual Orientation. Atlanta: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

65.

Waterman

Dawson

Bologna

1989. “Sexual Coercion in Gay Male and Lesbian Relationships: Predictors and Implications for Support Services.” Journal of Sex Research 26:118–24.