Abstract
Two hundred forty-seven (N = 247) undergraduate women at a medium-sized, Southwestern university provided written descriptions of a hypothetical sexual assault (SA). Women with a prior history of SA also described their actual SA experiences; women without a SA history provided a written description of a prior bad date or hookup. The contextual features of SA scripts were compared to those of actual SA experiences. Several characteristics of a stereotypical or “blitz rape” (e.g., physical force by a stranger) were more likely to be included in SA scripts relative to women’s actual SA experiences. Victimized women were also more likely to include verbal coercion, a hangout/hookup context, and previous consensual kissing in their SA experiences, in comparison to their SA scripts. The contextual features of SA experiences were also compared to the contextual features of bad dates or bad hookups. SA experiences, relative to bad dates, were more likely to include alcohol use, physical and verbal coercion by the perpetrator, and passive resistance. SA experiences, relative to bad hookups, were more likely to include physical and verbal coercion by the perpetrator, and knowing the man for less than 1 week. Victimized participants SA experiences were also found to be less likely to include previous consensual kissing and consensual intercourse in comparison to bad hookup experiences of nonvictimized women. Overall, there was considerable overlap between the contextual features present across all experiences. The lack of differentiation among these events may explain why women experience difficulty acknowledging whether they have experienced SA.
Introduction
Sexual script theory, developed by Simon (1973), is predicated on the concept of scripts as cognitive models used to appraise and navigate social interactions (Rose & Frieze 1993; Simon & Gagnon, 1986). Scripts involve both descriptive (e.g., contextual features and events that typically occur within a given social interaction) and normative elements (e.g., knowledge of appropriate and culturally accepted behaviors) (Krahé et al., 2007). Scripts evolve across development, as cultural influences lead to both observational and direct learning that is reinforced subsequently by peer groups (Frith, 2009).
The traditional heteronormative sexual script is the most frequently endorsed sexual script in the United States. Men are perceived as active sexual participants who are expected to have a high libido, initiate sexual activity, engage in casual sex with no commitments, and strategically overcome women’s resistance rather than accept rejection (Byers, 1996; Seal et al., 2008). By comparison, women are viewed as submissive and characterized as having little interest in sex except within a committed relationship, and as being responsible for resisting sexual advances to avoid social consequences (Byers, 1996). Gender differences in socially acceptable behavior may place women and men at cross-purposes in effectively negotiating sexual interactions. For example, a man may be more likely to engage in sexual aggression to override a woman’s resistance if he believes her resistance is not genuine or is “token resistance” (Canan et al., 2018), whereas a woman may be more likely to deny men’s sexual advances and be less likely to initiate sexual activity, even if interested in doing so (Muehlenhard & McCoy, 1991). However, sexual scripts may not be accurate representations of normative sexual interactions between men and women.
While scripts may not reflect lived experience, they can influence how people approach sexual interactions (Simon & Gagnon, 1986). Thus, sexual scripts may operate similarly to heuristics, or as cognitive shortcuts that do not effectively utilize available information and save effort with the cost of reduced accuracy of decision-making (Tversky & Kanheman, 1974). Likewise, scripts have been theorized to affect both attention and memory for context in social interactions (Baldwin, 1992; Schank & Abelson, 1995). Additionally, people may be less comfortable acting against perceived norms, as informed by scripts, which sets the stage for women to exercise less sexual assertiveness, thereby reducing their sexual self-efficacy in sexual interactions (Rickert et al., 2002).
Sexual scripts also encompass specific scripts about sexual assault (SA), which are associated with women’s risk for SA (Turchik et al., 2009; Yeater et al. 2020). SA is a major public health problem, one associated with numerous negative psychological and physical mental consequences (Classen et al., 2005; Faravelli et al., 2004; Kuehn, 2011; Ullman & Brecklin, 2003). To date, research has yet to fully investigate differences among SA scripts, real-life experiences of SA, and related, but disparate, negative interpersonal events. Explicating the differences among scripts and these real-life events may elucidate novel risk factors for SA.
Purpose of Current Study
The aim of the current study was to examine the congruence between (a) previously victimized women’s SA scripts and their SA experiences, and (b) victimized women’s SA experiences with nonvictimized women’s bad date or bad hookup experiences. To date, there is limited research on the contextual features that differentiate these events. In the following, we review the literature on SA scripts, hookup scripts, and date scripts and the similarities and differences among them; discuss script elements as risk factors for SA; and highlight the gaps in the literature this study attempts to bridge. We show that the contexts in which these events occur overlap, perhaps explaining partially why women experience difficulty recognizing that they have experienced SA. To facilitate ease of interpretation throughout this paper, “script” will be used when discussing women’s SA scripts, and “experience” will be used when discussing women’s real-life experiences.
SA Scripts
Koss (1985) defined rape as “oral, anal, or vaginal intercourse against [one’s] will through the use of force or threat of force” (p. 196) after a thorough examination of each state’s legal definitions of rape, the goal being to make the definition consistent with most states’ legislation. In contrast, Koss et al. (1987) defined SA by a range of behaviors increasing in severity, from unwanted sexual contact to completed rape (Koss et al. 1987). Using these definitions, Koss et al. (1987) found that with 25% of college women reported attempted or completed rape during their college years (Krebs et al., 2007), and 90% of perpetrators of SA are men (Basile et al., 2011).
Adherence to stereotypical SA scripts (i.e., holding stereotypical beliefs about what commonly occurs during a rape, including sequence of events, location, level of force) is thought to increase the probability of women experiencing SA by interfering with their capacity to recognize risks that do not fit their scripts (Turchik et al., 2009; Yeater et al. 2020). For instance, Stirling et al. (2020) indicated 44 to 47% of participants between the age group of 18 and 25 years believed in a “blitz rape” or a “real rape,” a script in which a stranger, often in possession of a weapon, attacks a woman, threatening physical force, and forcibly rapes her, typically at night or in an isolated location outdoors (Kahn et al., 1994; Littleton & Dodd, 2016). Despite the pervasiveness of the “blitz rape” script, results from the 2010 National Crime Victims Survey estimated that only 12% of SA involved a weapon and only 8% of SA involved a stranger with no prior relationship to the victim (BJS, 2011), suggesting that this type of SA is very uncommon. In fact, 78% of SAs were perpetrated by acquaintances, friends, or intimate partners of the victim (BJS, 2011). Women whose script dictates that strangers are the most likely to assault them may not recognize signs of risk in real-life settings with people who are known because those signs are discordant with their beliefs about what is likely to occur in SA (Turchik et al., 2009; Yeater et al., 2020). Similarly, research shows that 95.3% of SA did not involve physical force (Krebs et al., 2007), 89.2% of SA victims reported drinking prior to the incident (Krebs et al., 2007), and 70.1% of victims reported the assailant was drinking prior to the incident (Krebs et al., 2007), all factors that are inconsistent with the “blitz rape” script. Script elements such as nonforceful resistance, an outdoor setting, a higher severity of assault, and the woman having known the perpetrator for a shorter time period were predictive of women reporting a SA over the academic term, suggesting that these women may have difficulties detecting evidence-based risk factors (Turchik et al., 2009). Further, Yeater and colleagues (2020) found that a more severe sexual victimization history, a putative risk factor for SA, was associated with higher endorsement of the Stereotypical Assault Scripts subscale of the Sexual Assault Script Scale (SASS). The SASS is hypothesized to measure participant’s beliefs about the presence of certain elements of SA, and the Stereotypical Assault Scripts subscale contains several items that correspond to the “blitz rape” script (e.g., How likely is it that he overpowers you physically? How likely is it that you are in an alley?).
Less than half of women whose SA meets the legal definition of rape acknowledge their experience as such (Wilson & Miller, 2016). The lack of SA acknowledgment is posited to be a result of the discrepancy between women’s SA scripts and their SA experiences (Littleton & Axsom, 2003; Ryan, 2011). In other words, women may not recognize their experiences as constituting SA because their experience is not consistent with what they think happens in a typical SA, and this discrepancy may heighten risk for SA revictimization (Littleton et al., 2017).
While women’s SA scripts may be associated with their risk for SA, this does not mean that women are to blame for SA, as that responsibility lies solely with the perpetrator. However, identifying mutable risk factors and understanding how they may affect overall risk of SA may inform the development of interventions for women that may assist them in managing these risks in real-life situations. This perspective is especially relevant given the paucity of existing effective interventions for reducing men’s propensity to engage in sexually aggressive behavior. Indeed, to date, very few interventions have proved successful at ameliorating this important public health problem (Ellsberg et al., 2015).
Hookup Scripts
Hooking up is defined as a brief, uncommitted sexual interaction among individuals who are not dating (Garcia et al., 2012). Hookups, or spontaneous sexual encounters with little to no commitment, are a subset of sexual scripts that have become increasingly prevalent among emerging adults between the age group of 18 and 25 years (Garcia et al., 2012; Olmstead et al., 2019). Holman and Sillars (2012) found ten themes of hookup scripts (e.g., attending parties, flirting/attraction, dancing, and friend arrangement), six of which were previously identified by Paul and Hayes (2002). The three most dominant themes were spontaneous behavior, friends or acquaintances being present, and alcohol consumption. Many elements of hookups were inconsistent across participants’ accounts (e.g., the type of sexual activity varied from kissing to intercourse and the relationship between partners ranged from stranger to nonromantic friend), but spontaneous behavior emerged as a key feature of hookups (Holman & Sillars, 2012; Paul et al., 2000). Notably, casual sexual encounters such as hookups are posited to occur in contexts that confer significant risk for SA (Littleton et al., 2009). Littleton and colleagues (2009) found that 22% of assault victims reported that their assault initially started out as a hookup, 39% of rape victims reported the perpetrator was a date or a romantic partner, and 45% of assault victims reported that they had participated in binge drinking prior to the incident. Though women tend to view hookup contexts as disparate from the contexts in which rape occurs, due in part to the pervasiveness of the “blitz rape” script, hookups are still perceived as emotionally risky with the potential to cause negative psychological consequences such as shame (Littleton et al., 2009).
Date Scripts
The literature on date scripts primarily comprises of heteronormative scripts for a first date. In this category, Rose and Frieze (1989) found that scripts for women focused on the private sphere (e.g., consideration of appearance, conversation, and controlling their sexuality) and the scripts for men frequently involved the public domain (e.g., planning and paying for the date), and recent research suggests that these gendered expectations have remained consistent (Cameron & Curry, 2020). However, participants who were women, who described a date between friends, and who endorsed less sexist attitudes were more likely to endorse egalitarian scripts.
Related research has examined the contextual features of dating experiences, including those that end in SA (Muehlenhard & Linton, 1987; Yeater et al., 2008). In their 1987 study, Muehlenhard and Linton identified how persistent rape myths and misconceptions about what a typical rape looked like impeded women’s efforts to recognize and respond to rape attempts. While identifying risk factors for SA, Muehlenhard and Linton (1987) also examined the contextual features of sexually aggressive dates relative to dates that involved no sexual aggression. They found that heavy alcohol use, “parking,” and a party setting were associated with sexual aggression on dates. Yeater and colleagues (2008) explored whether contextual features of sexually aggressive dates had changed since the initial study and found that women were more likely to experience sexual aggression when the man had consumed alcohol or used drugs, the woman reported alcohol use and experienced difficulty being assertive in the dating situation, and the date started with a larger group setting. These findings were consistent with Muehlenhard and Linton’s (1987) finding of the association between sexual aggression and a party context.
Limitations of Past Research
Examining women’s SA scripts appears relevant in better illuminating women’s SA risk, as previous work has suggested that women’s SA script characteristics (e.g., nonforceful resistance, the assault occurring outdoors, and the assault being more severe) are associated with putative risk factors for SA (Yeater et al., 2020) and prospectively predicted SA experiences (Turchik et al., 2009). However, to the authors’ knowledge, there has been no research investigating the discrepancy or concordance between women’s written descriptions of their assault experiences and their SA scripts. Thus, the present study may help to identify specific subgroups of women who may be at an increased risk for SA, such as women whose beliefs adhere to more stereotypical scripts of SA and who may not consider acquaintances, friends, or romantic partners as potential perpetrators. Furthermore, examining the relationship between women’s SA scripts and real-life SA experiences could elucidate potential explanations regarding why women at heightened SA risk (e.g., women who report a prior SA history) may have difficulty identifying risk (Yeater et al., 2010, 2020). The present study addresses an additional gap in the literature through its examination of the correspondence between SA experiences and accounts of a “bad hookup” or a “bad date.” Previous studies have examined the contextual features among dates that did or did not involve sexual aggression (Muehlenhard & Linton, 1987; Yeater et al., 2008), and Littleton et al. (2009) specifically compared elements of college students’ SA scripts with elements of their “bad hookup” scripts. By limiting hookup and date experiences to those that specifically included negative experiences, the present study aimed to discriminate further between SA and related, but presumably different, negative interpersonal events. Through a qualitative examination of SA scripts, actual SA experiences, and actual bad date or bad hookup experiences, we aimed to discern the similarities and differences in contextual features among each type of event.
In addition, because many survivors of SA do not identify their experiences as “rape,” extant research on SA scripts has obtained a restricted range of responses by using the word “rape” in the instructional sets, unintendedly excluding those with an unacknowledged SA (e.g., Littleton & Axsom, 2003; Littleton et al., 2009). Thus, SA scripts such as sexual coercion, alcohol-facilitated rape, and attempted rape were less likely to be described by participants. To account for the fact that many participants may have different perceptions of behaviors that constitute SA, Turchik et al. (2009) and Leiting and Yeater (2017) used a broad definition of a SA in the instructions to participants and asked several specific questions about the context, location, and relationship to the man. The present study’s comprehensive definition of SA (i.e., usage of the Sexual Experiences Survey to identify participants with a victimization history) and inclusion of specific, contextual questions were intended to generate more complete descriptions of each type of event.
Goals of Current Study
As noted, the aim of the current study was to examine the correspondence between (a) previously victimized women’s SA scripts and their SA experiences, and (b) victimized women’s SA experiences with nonvictimized women’s bad date or bad hookup experiences. Littleton et al., 2009 compared and contrasted themes and outcomes of participants’ SA scripts and bad hook up scripts; thus, we extended prior work by examining the differences between real-life experiences with SA and bad dates or bad hookups. We expected that SA scripts, relative to SA experiences, would include many features typical of the “blitz rape,” yet not common features of SA perpetrated by acquaintances. Given the lack of literature on the differences between bad dates or bad hookups and SA experiences (Littleton et al., 2009; Yeater et al., 2008), we analyzed bad date and bad hookup experiences separately rather than collapsing them into one category. We anticipated that women’s SA experiences would include more physical and verbal coercion by the man, active resistance by the woman, and increased alcohol consumption relative to nonvictimized women’s bad date or bad hookup experiences.
Method
Participants
Data were collected from 252 ethnically diverse undergraduate women enrolled in psychology courses at a medium-sized university located in the southwest region of the United States. As part of a larger, previously published study (Leiting & Yeater, 2017), participants were recruited through the psychology department’s research pool and received course credit for their participation in the study. As most women with a history of SA report first experiencing completed or attempted rape before the age of 25 (Smith et al., 2018), women outside of this age range were excluded from participation. Participants’ mean age was 19.28 (SD = 1.45), with most participants reporting being single (89%, n = 220), heterosexual (98%, n = 248), and either freshman (51%, n = 126) or sophomores (22%, n = 54). Most participants self-identified as Hispanic/Latina (48.2%, n = 119) or White/Caucasian (37.2%, n = 92). Since the study focused on women who were interested in having dating and sexual relationships with men (who are the common perpetrators of SA), women were excluded from the analysis if they self-identified as lesbian/gay or reported being married (n = 5); thus, the current analysis included data from 247 undergraduate women.
Procedure
The study was conducted in compliance with the university’s Institutional Review Board. Upon arriving at the lab, participants were met by a research assistant who provided them with information about the study, obtained informed consent, and gave instructions on study procedures. Participants were informed that they could withdraw at any point during the study by informing the researchers of their intent to end their participation. No participants chose to withdraw during either the informed consent or study procedures.
Participants first completed the SA script by writing a description of a situation in which they imagined being verbally or physically coerced by a man into a sexual experience. Participants were asked to answer the following questions in their written script: Please describe what happens before, during, and after the situation. In addition, please answer the following questions: Who is the man? How long have you known him? What is your relationship to him? Describe where you are and what you are doing. What are you and the man thinking? What activities will you be engaging in? What type(s) of sexual activity is occurring? How does the situation end?
This script was adapted from previous studies (Nurius et al., 2000; Turchik et al., 2009) and was modified to include follow-up questions that probed for additional details. After participants completed their SA script, they completed self-report questionnaires inquiring about demographic information and their SA history. If a participant reported prior SA on the Sexual Experience Survey (SES), they were asked to write about their most severe SA experience based on the highest numbered item they endorsed on the SES; if they did not, they were asked to write about a bad date (a planned social activity) or bad hookup experience (an unplanned, spontaneous social activity involving sexual contact). Previous qualitative studies have left the definition of “bad” experiences up to the individual participant in order to elicit a wide range of negative consensual experiences (Littleton et al., 2009); thus, we followed the approach used in prior work. The same instructional set was given to participants to describe their actual experience as was used as in the SA script prompt. Participants were given 15 to 20 minutes each to complete the SA script and description of an actual experience (i.e., SA experience or bad date/hookup experience). Upon completion of the study materials, participants were provided with a debriefing form, which included counseling resources if they experienced any adverse reactions due to participation. Participants took approximately 1 hour to complete all study procedures.
Self-Report Measures
Demographic Questionnaire
This self-report measure assessed participants’ age, sexual orientation, marital status, race/ethnicity, and academic status.
Sexual Experiences Survey
The SES is a 10-item self-report measure that assesses severity of SA (unwanted sexual contact to completed rape) since the age of 14 using behavioral descriptions of SA (Koss et al., 1987). Here is a sample item from the scale: “Have you had a man attempt sexual intercourse (get on top of you and insert his penis) when you didn’t want to by threatening or using some degree of force (twisting your arm, holding you down, etc.) but intercourse did not occur?” The SES has been found to have good internal consistency (α = .74), 1-week test–retest reliability (r = .93), and a correlation of r = .73 with interview responses (Koss & Gidycz, 1985). Using a nonredundant categorization by assault type severity (Koss et al., 1987), a common scoring technique for the SES (Davis et al., 2014), approximately 23.5% (n = 58) of participants reported no victimization, whereas approximately 65.6% (n = 162) of participants reported SA since the age of 14. Of the victimized participants, 23% (n = 54) reported moderate victimization (i.e., unwanted sexual contact or sexual coercion) and 44% (n = 108) reported severe victimization (i.e., attempted or completed rape).
Coding System: Categorization
To ensure that our categorizations of the written experiences (i.e., SA, bad dates, and bad hookups) were accurate, we manually reviewed each written description for accuracy rather than relying solely on participants’ responses to SES items. Given that over half of women who have experienced SA do not label it as such (termed “unacknowledged rape”) (Littleton & Axsom, 2003; Wilson & Miller, 2016), it seemed particularly relevant to check the descriptions provided by participants to ensure that they were categorized accurately (i.e., SA vs. bad date or hookup).
Two individuals coded the contextual features of each description, and disagreements in codes were resolved in discussions with the larger research team (see description of coders in the following paragraph). When an experience was ambiguous, and we were unable to classify them due to a vague or incomplete written account (e.g., a participant described an initially consensual sexual experience, but it was unclear from the description if it became nonconsensual), we retained the original categorization of SA or non-assault according to the SES items. We recategorized participant’s experience if (a) the description described a SA, yet the participant did not endorse any item on the SES; or (b) the description did not describe a SA, but the participant reported that they experienced a SA on the SES (e.g., instead of describing the SA, they described a nonsexual assault event). If the latter occurred, we removed the data from the analyses, as the participant did not appear to follow the experimental instructions, leading to concern about the integrity of their data. Because of this process, 130 experiences were coded as SA, 22 were coded as bad dates, and 17 were coded as bad hookups. Twenty (N = 20) women provided written descriptions that were recategorized as SA, whereas 58 participants were removed from the analysis for (a) having selected an SES item but failing to provide a written description of that event (e.g., instead describing a bad date or hookup); (b) missing data from the SES and a written description that was sufficiently ambiguous that it was impossible to categorize the event; or (c) not describing any of the three categories (i.e., SA, bad hookup, or bad date).
Coding of the Contextual Features of the Scripts and Experiences
The original coding system for the written descriptions was created inductively (e.g., using a “bottom-up” approach) to ensure that the themes identified by the coding system were related to the data. The descriptions were inspected to identify thematic and contextual similarities and differences; codes (and coding system) resulted from this process. Thirteen general categories emerged: (a) alcohol use, (b) drug use, (c) location (e.g., his property and her property), (d) relationship to the perpetrator (e.g., acquaintance and coworker), (e) previous consensual sexual contact, (f) active resistance (e.g., yelling/screaming and physically resist), (g) nonforceful resistance (e.g., crying and saying no), (h) verbal coercion, (i) physical coercion, (j) situation context (e.g., date and afterparty), (k) length of time the woman had known the perpetrator, (l) victimization severity, and (m) negative psychological effects. Specific codes were developed to capture the data within each category (e.g., “feels dirty or disgusting” and “blames self” were specific codes within the general category of “negative psychological events”). Coders included a professor who is an expert in the sexual violence research area and four female graduate students and one male graduate student, all with extensive experience in SA research. Coders were blind to participants’ self-report measures, including demographic information and history of SA. Coders were trained on the coding system using example descriptions, and upon reaching an interrater reliability greater than a kappa value of .70, they coded the actual participant descriptions (see Leiting & Yeater, 2017, for more information on the training procedures for coders). The mean interrater reliability kappa value was .85, and kappa values ranged from .79 to .87. Given all kappa values were above .70, the coding system was considered to have acceptable interrater reliability.
Data Analytic Strategy
Descriptive statistics were analyzed to determine the frequencies of each contextual feature within each coding category. Chi-square analyses then were used to examine differences in contextual features between victimized women’s SA experiences and their scripts and between victimized women’s SA experiences and nonvictimized women’s bad date or bad hookup experiences. SA experiences were collapsed into two categories (SA or no SA), since examining all SES categories using chi-square analyses resulted in several unreliable results due to expected cell counts below 5.
Results
Victimized Women’s Scripts and SA Experiences
Several significant differences in script characteristics emerged between victimized women’s hypothetical SA scripts and their actual SA experiences. Victimized women were less likely to report the following in their SA experience relative to their SA script: (a) both the man and the woman drinking alcohol (17% vs. 32%, X2[1, N = 150] = 3.990, p = .046); (b) knowing the man for a less than 1 week (13% vs. 41%, X2[1, N = 150] = 7.821, p = .005); and (c) the man grabbing or pushing the woman (39% vs. 48%, X2[1, N = 150] = 3.841, p = .050). Victimized women were more likely to report the following in their SA experiences relative to their hypothetical SA scripts: (a) the man using verbal coercion (55% vs. 47%, X2[1, N = 150] = 7.406, p = .007); (b) the context of a hookup/hangout (25% vs. 13%, X2[1, N = 150] = 15.504, p < .001); and (c) previous consensual kissing (48% vs. 30%, X2[1, N = 150] = 4.455, p = .035).
A representative example of a victimized women’s SA script and a victimized woman’s SA experience, respectively, follows.
SA Script Example
The man who I imagine to verbally or physically coerce me into a sexual experience is a stranger. I was walking back to my house, and it was getting dark so around 7:00pm or so. I saw that guy getting me by the back and then he knocks me out. Before I get knocked out, I am confused and then I get scared and panic. The man has probably planned all this. When I wake up, I am locked in a dark room. I am laying down with my hands and feet tied. The man enters the room and gives me a cup of water. I refuse to drink because it may have a drug. What I would have probably thought at that moment is that he wants to rape me then kill me. The man makes me drink the water. I finish all the water and then begin to feel dizzy. I imagine the man to do horrible stuff to me then rapes me, having sexual intercourse. I fall asleep and wake up in some place I don’t know.
SA Experience Example
We had been dating for over six months. He was manipulative and controlling. We were hanging out alone at his apartment and he was getting fidgety. He came and sat down behind me and wrapped his arms around me. He started playing [with] my boobs through my shirt and kissing my neck. He then took off his shirt and pulled mine off. He tried to put his hands down my pants, but I stopped him. So, all of a sudden he unbuckled his pants, and he grabbed my hand and started to [pull it] down. I was hesitant to do so because of the fact that I did not want have sex, but the fact that he was older and intimidating. I gave in. He wanted me to please him, so reluctantly I did. He ended up being kind of rough (biting, pulling my hair, etc.).
Victimized Women’s Assault Experiences and Nonvictimized Women’s Bad Date or Bad Hookup Experiences
The analysis between SA experiences and bad date experiences found that victimized women were more likely to report the presence of alcohol (36% vs. 5%, X2[1, N = 172] = 8.727, p = .003), verbal coercion by the man (55% vs. 14%, X2[1, N = 172] = 13.343, p < .001), physical coercion by the man (77% vs. 41%, X2[1, N = 172] = 12.818, p < .001), and passive resistance by the woman (49% vs. 23%, X2[1, N = 172] = 5.209, p = .022) relative to nonvictimized women’s date experiences. Nonvictimized women’s bad date experiences were more likely to be in the context of a date (59% vs. 9%, X2[1, N = 172] = 35.895, p < .001), in comparison to victimized women’s SA experiences, which was expected given that we asked nonvictimized participants to describe a date.
Overall, there were several significant differences in contextual factors between SA and bad hookup experiences with victimized women being more likely to report verbal coercion by the man (55% vs. 12%, X2[1, N = 167] = 11.598, p < .001) and physical coercion by the man (77% vs. 35%, X2[1, N = 167] = 13.709, p < .001), relative to nonvictimized women’s hookup experiences. Nonvictimized women’s bad hookup experiences were more likely to describe previous consensual kissing (76% vs. 48%, X2[1, N = 167] = 4.952, p = .026), previous consensual intercourse (24% vs. 2%, X2[1, N = 167] = 17.624, p < .001), and knowing the man for less than 1 week (35% vs. 14%, X2[1, N = 167] = 5.036, p = .025) as compared to victimized women’s SA experiences.
An example of a nonvictimized woman’s bad date experience and a bad hookup experience follows.
Bad Date Experience Example
I met the guy that I have been with while we were in high school. I described a bad date that I had with him. Well one time he was supposed to take me to the movies. We went out to eat at a Chinese restaurant first and he started drinking sake. After the Chinese restaurant he was too buzzed to drive so I had to drive us to the movies. Then while we’re at the movies, he fell asleep during the entire movie. After the movie was done and we were heading home, on the way home he started kissing my neck [like] his previous activities had no effect on me at all. We get back home and get into the house, and he starts rubbing and hugging [me]. I thought this was outrageous. How dare he think he was going to get some after acting like that the entire day. He continued to try to rub and kiss me, after 15 min past [sic] and me being tired of his sexual desires, I then turned around and smacked him right against the head. I said no means no. He was pretty upset, he rolled over and went to sleep. When he woke up the next morning, he was reminded of his stupid actions. He apologized and learned that those type of actions would be strictly forbidden if he wanted to be with me.
Bad Hookup Experience Example
I hooked up with this guy [who] was my cousin’s friend. I had known him for about three years, and we would talk on and off. [During] my junior year in high school I had gotten out of a bad relationship and my cousin’s friend started talking to me. We would text every day, this occurred for about two months. I was house sitting for my uncle and I had some friends over to hang out. I invited him over. He showed up and we all hung out for a bit [and] my friends ended up going to sleep. We talked for a bit and then we ended up having sexual intercourse. After we were done, we realized the condom had broken but I didn’t really worry too much until the morning. I went to Walgreens and got Plan B. We did not really talk much after but when we see each other we will talk and say hi.
Discussion
The purpose of the current study was to investigate the congruence between victimized women’s SA experiences and their SA scripts, and the congruence between victimized women’s SA experiences and nonvictimized women’s bad dates or bad hookups. Our findings extend the literature in several ways, which we discuss in the following.
Features of SA Scripts in Women with a Prior SA History
Our findings revealed that while victimized women do include some characteristics in their SA scripts consistent with SA statistics, they still include some features that are relatively uncommon characteristics of SA experiences and that are also inconsistent with their own SA experiences. Such features included knowing the man for less than a week (BJS, 2011), and the man using physical force such as grabbing or pushing (Krebs et al., 2007). This is particularly concerning as similar contextual script features have been found to predict SA longitudinally (Turchik et al., 2009). Previously victimized women additionally described a hookup context and verbal coercion by the man more frequently in their actual SA experiences than in their scripts. These findings suggest that college women, on average, do not think, commonly, about SA occurring within the context of a familiar relationship or where the man might use continual arguments to pressure a woman into sexual activity, despite the disproportionate role these circumstances play in SA (Fielder et al., 2014; Testa et al., 2010).
Unexpectedly, victimized women also included alcohol use more often in their hypothetical scripts than in their actual SA experiences, even though previous research has found that alcohol use is frequently a part of SA (Testa et al., 2010; Tyler et al., 2017). This may be due to the relatively nontraditional college population at the university where the study was conducted, which includes a substantial group of students who live off campus or commute (93%), as well as students who are 25 years or older (22%) (University of New Mexico Office of Institutional Analytics, 2022). Thus, these women’s SA experiences may have been atypical; that is, relative to those who are members of more traditional college student populations in which alcohol use and parties occur more frequently and in which Greek life is a central feature of the university environment and culture.
Features of SA Experiences Compared to Bad Date or Bad Hookup Experiences
Women who reported SA, bad dates, and bad hookups reported similar emotional experiences of feeling dirty or disgusting afterward, being ashamed or embarrassed, blaming oneself, feeling bad or guilty, feeling angry about the event, and feeling numb after the event. That is, there were no statistically significant differences among these experiences with respect to negative emotional outcomes. Future research should endeavor to better understand these outcomes and focus on understanding how to enhance women’s ability to have positive sexual experiences, even if those experiences might involve casual (and desired) sexual encounters. While prevention programs for women often focus exclusively on reducing behaviors that might heighten SA risk, we believe that interventions that enhance sex positivity among women, while also focusing on how to reduce SA risk (i.e., increase positive sexual outcomes while reducing the negative ones) might be a promising avenue for intervention science.
Of course, it is possible that some bad date or bad hookup experiences reflected unacknowledged SA, which research suggests occurs in over half of SA incidents (Littleton et al., 2007; Rousseau et al., 2020), and yet, these women did not report SA on the SES. It may be that women whose bad date or hookup experiences resulted in these negative feelings are unacknowledged victims of SA, which may suggest that we consider alternative methods for querying women about their SA experiences. Alternatively, bad date and hookup experiences may intersect with SA in currently unexplored ways. Thus, future research might explore the ways in which these events differ contextually and emotionally.
Notably, there also were contextual differences between victimized women’s SA experiences and nonvictimized women’s bad date or hookup experiences. Victimized women’s SA experiences more frequently described verbal and physical coercion by the man relative to nonvictimized women’s bad date or bad hookup experiences, which is consistent with previous research (Kern & Peterson, 2020). Victimized women’s SA experiences were also more likely to include active resistance by the woman, relative to nonvictimized bad date experiences. This is consistent with research findings that 84% of rape victims physically resisted or tried to physically resist, 64% tried to escape or get way, and 92% said “no” or “stop” (Cantor et al., 2021).
Unsurprisingly, nonvictimized women’s bad date experiences more frequently occurred in the context of a date in comparison to victimized women’s SA experiences. Similarly, it seems that nonvictimized women may be hooking up with previous partners repeatedly, as previous consent to engage in kissing or intercourse were more likely in bad hookups compared to SAs, consistent with prior work (Littleton et al., 2009). Women who wrote about SA experiences were more likely to include alcohol use than women who wrote about a bad date. The presence of alcohol and potential for intoxication in SA experiences is consistent with the literature showing that alcohol intoxication reduces women’s ability to recognize risky situations and resist unwanted sexual advances (Davis et al., 2009). Future research may consider examining the frequency of alcohol use within the context of bad hookups, as previous literature has found that at least half of students expected alcohol to be involved in hypothetical bad hookup scenarios (Littleton et al., 2009).
Contextual findings suggest that women may benefit from learning different strategies for responding to types of high-risk situations, with hookup situations being identified as a type of risky situation that varies in identifiable ways from a SA situation. Intervention efforts may focus on helping women learn techniques to achieve their goals (a consensual sexual experience), while also learning ways to mitigate risks associated with hookups. Additionally, given the negative emotions associated with hookups in the study’s sample, and the potential for unacknowledged rape, attending to the signs of sexual coercion and ways to self-advocate in sexual situations may also be especially relevant (Rousseau et al, 2020).
Incongruence Between SES Endorsement and Recategorized Experiences
A considerable number of women in the study did not endorse items on the SES, yet their written descriptions described clearly incidents of SA as determined through our coding process (N = 20). It is interesting to note that even though victimized women, as expected, described physical coercion significantly more than nonvictimized women, 38% of nonvictimized women described physical coercion (e.g., grabbing, touching, and pulling) on the part of the man during their bad date or bad hookup experience. It is possible that despite attempts to categorize participant’s experiences based on their written descriptions; these women did in fact experience sexual violence that was not explicitly described in their responses. Related work supports such a possibility, as there may be several reasons why women in this study did not identify as victims of SA, if these events were in fact SA experiences. Rousseau and colleagues (2020) found that adherence to rape scripts and rape myths, ambivalence about labeling a SA as such, social or emotional consequences that may result, or lack of knowledge about consent and appropriate sexual behaviors may be associated with not identifying as a victim of SA. Thus, while Koss and colleagues (Koss & Gidycz, 1985; Koss et al., 1987) revolutionized the measurement of SA by focusing on behavioral descriptions of SA that avoided terms such as “rape,” ultimately leading to more accurate prevalence rates of SA, our current methods may still be underestimating the number of women who have experienced SA.
Limitations and Future Directions
There are limitations to the current work. First, women’s accounts of their SA experiences and bad date or bad hookups were based on retrospective report. The use of methods such as ecological momentary assessment would facilitate a more accurate understanding of the situational precipitants that precede SA in near real time (Yeater et al., 2022). Furthermore, the instructional prompt did not provide participants with a functional definition of a “bad” date or hook up experience, thus participants may have interpreted this word differently. While this methodology is consistent with previous work (Littleton et al., 2009), future studies examining contextual differences may consider using a more specific definition to narrow the range of responses. Second, because this study was cross-sectional in design, we cannot determine whether the contextual features identified in women’s hypothetical SA scripts predict SA. Yet prior work has identified that certain features found in women’s hypothetical SA scripts in this study do predict SA prospectively (Turchik et al. 2009) and are associated with putative risk for SA (Yeater et al., 2020). Third, collapsing women’s SA experiences into one group rather than multiple groups that reflect disparate SA severity may have obscured important differences in the contextual features in these experiences. Finally, while the study sample was quite diverse racially and ethnically, the generalizability of our findings should be examined in more diverse groups of women, including (but not limited to) young women who do not attend college, women who identify as gay/lesbian, and college women in disparate U.S. areas and other countries.
Despite the fact that SA has become a relatively common topic in our culture (e.g., #MeToo movement), a “blitz rape” is still a normative script held among college women. In addition, the fact that lay definitions of rape often differ from legal definitions (Haugen et al., 2018), indicates that our culture still holds inaccurate assumptions about what constitutes sexual violence against women. While change is greatly needed at a societal level, a short-term solution may be for preventative interventions to focus on providing at-risk women (i.e., previously victimized women) with personalized and normative feedback about how they construe SA, given the study’s finding that SA scripts are inconsistent with features of real SA (Littleton & Dodd, 2016; Littleton et al., 2009). Personalized feedback interventions have been used successfully with college students to reduce sexually risky behavior and alcohol consumption (e.g., Kilwein et al., 2017; Lewis et al., 2014). This type of an approach might reduce women’s adherence to more stereotypical rape scripts, which then might increase their ability to orient to environmental cues that signal risk for SA. Of course, a priority research target should be changing men’s sexually aggressive behavior since most sexually violent crime is committed by men. However, given interventions focused on changing male aggression have been largely unsuccessful, assisting women in making sexual decisions that are consistent with their values, and their sense of sexual agency may prove useful in reducing risk in the meantime.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interests with respect to the authorship and/or publication of this article.
Funding
The authors received no financial support for the research and/or authorship of this article.
