Abstract
We examined the role of sexting coercion as a component of the intimate partner abuse (IPA) construct among young adults to determine whether sexting coercion would emerge alongside other forms of partner aggression as a cumulative risk factor for psychological, sexual, and attachment problems. In a sample of 885 undergraduates (301 men and 584 women), 40% had experienced some type of coercion. Although there was some overlap between sexual coercion and sexting coercion (21% of participants had experienced both), some individuals had experienced only sexting coercion (8%) and some only sexual coercion (11%). Women were more likely than men to be coerced into sexting. Both sexting coercion and sexual coercion were significantly and independently related to negative mental health symptoms, sexual problems, and attachment dysfunction, and, notably, sexting coercion was found to be a cumulative risk factor for nearly all of these negative effects. These data support the idea that digital sexual victimization is a new component of IPA polyvictimization, potentially increasing the negative effects experienced by victims of multiple forms of partner aggression.
Research into intimate partner abuse (IPA) typically delineates three forms of aggression: physical, sexual, and emotional/psychological (Follingstad & Bush, 2014). However, despite the conventional distinction, these three types of abuse often overlap, with many individuals experiencing multiple forms of abuse in their intimate relationships (e.g., Campbell & Soeken, 1999; Dienemann et al., 2000; Katz, Moore, & May, 2008; Marshall, 1996). The experience of multiple forms of abuse within close relationships is so common, in fact, that terms such as covictimization and polyvictimization have emerged in the literature to describe the phenomenon. For example, Smith, White, and Holland (2003) defined covictimization as “both physical and sexual assault occurring within the same time period but not necessarily simultaneously during a single violent event or with the same perpetrator” (p. 1104). The term polyvictimization is more often used by researchers examining all three forms of traditional IPA (i.e., physical, sexual, and psychological) or by those studying less traditional forms of aggression, such as stalking, statutory rape, or Internet harassment (e.g., Cuevas, Sabina, & Picard, 2010; Hamby, Finkelhor, & Turner, 2012; Sabina & Straus, 2008).
There is some evidence that polyvictimization represents a distinct type of abuse and is associated with worse outcomes (e.g., more relationship dissatisfaction and negative mental health symptoms) than the experience of a single type of abuse (Campbell & Soeken, 1999; Katz et al., 2008; Sabina & Straus, 2008). Moreover, polyvictimization has been associated with increased negative effects even though it is not reliably associated with more incidents of abuse or higher levels of severity (Katz et al., 2008). This suggests that it may be the experience of multiple forms of abuse that creates these additive negative effects, rather than the experience of more abusive incidents. Indeed, research into polyvictimization among youth, where this phenomenon has received relatively more attention, suggests that polyvictimization is associated with more negative outcomes than repeat victimization of a single type, and that studies focusing on just a single form of abuse fail to capture the full experience of victims (Turner, Finkelhor, & Ormrod, 2010). We argue that the same principle applies to adult IPA victimization, and that the definition of IPA may be too narrow in its current form.
In a recent study, Drouin, Ross, and Tobin (2015) suggested that sexting coercion (i.e., experiencing coercion from an intimate partner to send sexually explicit text messages) is a part of the IPA construct. In that mixed-gender sample, sexting coercion was related to both sexual coercion and intimate partner violence. Moreover, those who had experienced sexting coercion reported more symptoms of anxiety, depression, and trauma than those who had not. However, Drouin et al. (2015) did not consider the potential additive effects of polyvictimization (i.e., sexting coercion in addition to more traditional forms of IPA) on mental health symptoms. Thus, the goals of the present study include a closer examination of the relation between sexual and sexting coercion, as well as between coercion victimization and sexual behavior, and determining whether sexting coercion is associated with negative outcomes for victims, above and beyond what is reported by those who experience more traditional forms of IPA exclusively.
Intimate Partner Aggression
IPA can be very broadly or more narrowly defined, yielding varying statistics on its prevalence and findings with regard to correlates (Follingstad & Bush, 2014; Hamby, 2014; Robbins, Hurley, Liu, & Chao, 2015). Adolescents and young adults may be particularly at risk. One recent longitudinal study found that 40% of adolescents and young adults had experienced IPA in the past year (Halpern, Spriggs, Martin, & Kupper, 2009). Humphrey and White (2000) found that nearly 70% of female college students experienced at least one incident of sexual victimization (unwanted contact, verbal coercion, attempted rape, or completed rape) during high school or college. In another study, more than 80% of college students reported psychological abuse perpetration or victimization (Fass, Benson, & Leggett, 2008).
IPA victimization has been linked to a host of problematic relational and psychological effects, including sexual problems, attachment dysfunction, and depression and anxiety. In her review of 40 years of intimate partner violence and sexuality research, Coker (2007) found that the experience of IPA was related to a variety of sexual issues, including risky sexual behavior and sexual dysfunction. More recently, Hellemans, Loeys, Dewitte, De Smet, and Buysse (2015) found higher rates of sexual dysfunction and attachment issues among those with histories of IPA victimization. The adult romantic relationship may be viewed as a primary context for security and attachment (Hazan & Shaver, 1994), and problems in this relationship context, including IPA, may disrupt this security or attachment bond, creating further vulnerabilities (La Flair, Bradshaw, Mendelson, & Campbell, 2015). Dysfunctional adult attachment may manifest as attachment-related anxiety (e.g., fear of rejection or abandonment by attachment figure) or attachment-related avoidance (e.g., discomfort with and/or resistance to interpersonal closeness; see Fraley, Hudson, Heffernan, & Segal, 2015), and recent data suggest that IPA victimization may contribute to the development of dysfunctional attachment patterns (La Flair et al., 2015). The experience of IPA also has been linked to psychological problems, including depression and anxiety (e.g., Shorey et al., 2011; Winstok & Straus, 2014). Thus, those who experience IPA in any form are at risk for a number of relational and psychological problems; however, research suggests that those who experience multiple forms of abuse may be at an even higher risk of these negative effects (Campbell & Soeken, 1999; Katz et al., 2008; Sabina & Straus, 2008).
Sexual Coercion
Over the past few decades, the scope of sexual aggression has expanded to include the act of coercion into sexual activity, including that which occurs in the context of an established relationship. Coerced sexual experiences are reported more commonly than forcible sexual assault, and generally involve repeatedly begging or pressuring an individual who is resisting sexual activity until he or she consents, often to end the coercive behavior or preserve the relationship (Livingston, Buddie, Testa, & VanZile-Tamsen, 2004; Struckman-Johnson, Struckman-Johnson, & Anderson, 2003). Rather than the physical force traditionally associated with sexual assault, manipulation is more often used as a tactic in sexual coercion.
In coercive situations within relationships, one partner generally employs postrefusal persistence tactics such as repeated begging, continually trying to arouse the other partner, threatening to end the relationship or seek sex elsewhere, or attempting to elicit sympathy (Katz & Tirone, 2010; Livingston et al., 2004; Struckman-Johnson et al., 2003). Other forms of coercion also occur (e.g., trying to intoxicate the other partner to lower his or her inhibitions) but are less common (Katz & Tirone, 2010; Livingston et al., 2004; Struckman-Johnson et al., 2003). Research has shown that both men and women experience sexual coercion, although women in the United States report it at a higher rate, and the tactics used differ across genders (O’Sullivan, Byers, & Finkelman, 1998; Sprecher, Hatfield, Cortese, Potapova, & Levitskaya, 1994; Struckman-Johnson et al., 2003). One study found that, while the majority of men (58%) and women (78%) experienced some form of coercion or postrefusal persistence, nearly twice as many women reported being persistently begged or lied to; men were more likely to have had unwanted sex after being continually aroused or enticed (Struckman-Johnson et al., 2003).
Although most coercive tactics are generally of the less severe variety (e.g., begging), as with forcible sexual assault, coercion has negative effects. Victims of sexual coercion may find their relationship with their current partner damaged, and some experience long-term psychological (Livingston et al., 2004), sexual (Zweig, Crockett, Sayer, & Vicary, 1999), and relationship (Collibee & Furman, 2014) consequences. Negative consequences have been demonstrated among both male and female victims of sexual coercion (Collibee & Furman, 2014). The adverse impact of non-violent coercion was also emphasized in a study by Katz and Myhr (2008), who found that undergraduate women who had experienced sexual coercion also reported more psychological abuse and lower relationship satisfaction, sexual satisfaction, and sexual desire. Furthermore, sexual coercion often results in unwanted sexual activity (Livingston et al., 2004; O’Sullivan et al., 1998), which is, in turn, related to lower relationship satisfaction (Katz & Tirone, 2010). Considered together, these studies suggest that the experience of sexual coercion, even non-violent, verbal forms of coercion, relates to negative relational, psychological, and behavioral outcomes (e.g., engaging in unwanted sexual activity).
Sexting Coercion
“Sexting,” a portmanteau of “sex” and “texting,” refers to “the sending of sexually explicit messages or images by cell phone” and may colloquially include a broad range of similar behaviors, including sending explicit videos of oneself, or using Internet sites or apps such as Snapchat to transmit sexual content (Drouin, Vogel, Surbey, & Stills, 2013; Judge, 2012). Recent studies indicate that most young adults have “sexted,” typically with a romantic partner, and that sexting is generally not associated with a broader pattern of deviant behavior or other sexual risk behaviors (Drouin & Landgraff, 2012; Gordon-Messer, Bauermeister, Grodzinski, & Zimmerman, 2013; Samimi & Alderson, 2014; Temple & Choi, 2014). Much of the extant literature on the negative effects of sexting focuses on the legal ramifications regarding sexual images of minors, or embarrassment due to sharing of the images beyond the intended recipient (Albury & Crawford, 2012; Judge, 2012). In this sense, it has been established that sexting can result in negative emotional consequences due to subsequent bullying or a damaged reputation, but the emotional consequences of sexting itself, particularly if one has been coerced into the activity by his or her partner, remain largely unexplored.
The coercive aspect of sexting has, to date, mainly been addressed in discussion of minor-to-adult sexting, and not within the context of an otherwise legal interaction between two people in a consenting relationship (Judge, 2012). However, recently, Drouin and colleagues (2015) established that consensual intimate partners are using coercive tactics to solicit sexually explicit photos and videos from their partners. In their sample of college students, one fifth of young adults had experienced sexting coercion in either their current or most recent romantic relationship. Moreover, those who experienced the most sexting coercion experienced high rates of sexual coercion, intimate partner violence, and mental health and trauma symptoms (Drouin et al., 2015). Based on these results, the authors concluded that, like sexual coercion, sexting coercion appears to relate to a variety of adverse effects. Moreover, they suggested that sexting coercion should be considered part of the IPA construct. However, Drouin et al. (2015) did not examine the unique or cumulative contributions of sexting coercion to mental health and trauma symptoms. Thus, it remains unknown whether sexting coercion and sexual coercion overlap in their prevalence or whether sexting coercion contributes unique variance to mental health and trauma symptoms, above and beyond other forms of traditional sexual coercion.
Present Study
The goals of the present study were to assess the impact of sexting coercion versus sexual coercion on behavior and psychological distress and dysfunction, and to determine whether individuals who experienced sexting coercion in combination with other forms of IPA experienced more negative effects than those victims who experienced more traditional forms of IPA alone. Based on research showing that sexual coercion is related to sexual compliance (e.g., Katz & Tirone, 2010; Livingston et al., 2004; O’Sullivan et al., 1998), we predicted that those who experienced the most coercion would engage in the highest rates of unwanted but consensual behavior, and that this would apply to both sexting and sexual coercion. In addition, considering the well-established negative effects of polyvictimization (e.g., Hamby et al., 2012; Sabina & Straus, 2008; Turner et al., 2010) and recent research suggesting that sexting coercion be considered part of the IPA construct (Drouin et al., 2015), we predicted that individuals who experienced sexting coercion in addition to more traditional forms of IPA (i.e., physical and psychological aggression) would report more negative mental health symptoms, sexual problems, and attachment dysfunction than those who experienced traditional IPA alone. We expected that those who experienced IPA plus both sexual coercion and sexting coercion would exhibit the most distress and dysfunction.
Method
Participants
Participants were 885 undergraduates (301 men and 584 women) from a midsized, Midwestern university. Most were college freshmen (62%) or sophomores (25%), with an average age of 20.43 years (SD = 4.64). Most were Caucasian (83%), followed by Hispanic (5%), African American (5%), Asian (4%), and 3% who identified as biracial, Native American, or “other.” Half (51%) were in a “committed” romantic relationship at the time of data collection, and those with no history of any committed or casual romantic relationship were excluded from analyses.
Procedure
Participants were recruited for an online study via the subject pool for introductory psychology. All participants completed the consent form and anonymous survey online.
Measures
The survey included demographic questions (e.g., age, sex, race, and major), questions about physical sex and sexting-related behaviors (e.g., age of first sexting and intercourse, sexting or having sex when one did not really want to), sexual and sexting coercion victimization, IPA, mental health symptoms, and romantic attachment.
Sexting and Sexual Behavior
Participants reporting any history of sexting or physical sexual intercourse were asked the age at which they sent their first sexually explicit text or picture message and their age at first intercourse. Participants also reported, on a 6-point Likert-type scale (0 = never, 5 = very frequently), how often they had consented to sex when they did not really want to have sex (i.e., unwanted but consensual intercourse [UCI]) and how often they sent a sexually explicit text, picture, or video message when they did not really want to (i.e., unwanted but consensual sexting [UCS]).
Sexting Coercion
To measure whether and under what circumstances participants had been coerced into sexting by their current or most recent romantic partner, participants completed an adapted version of the 34-item Sexual Coercion in Intimate Relationships Scale (SCIRS; Goetz & Shackelford, 2010). The SCIRS was selected because it measures coercion in intimate relationships, distinguishes between different types of coercive acts (e.g., threats to leave the relationship or threats to withhold resources), and has been used to measure sexual coercion within North American young adult samples (Goetz & Shackelford, 2010). The adaptation allowed for both men and women to report on their partner’s sexting coercion acts (e.g., “My partner persisted in asking me to send him or her a sexually explicit picture or video, even though he or she knew that I did not want to.”) that occurred during the past 12 months, or, in the case of past relationships, during the last 12 months of the relationship. Response choices on the SCIRS are as follows: 0 = “act did not occur,” 1 = “act occurred 1 time,” 2 = “act occurred 2 times,” 3 = “act occurred 3 to 5 times,” 4 = “act occurred 6 to 10 times,” and 5 = “act occurred 11 or more times.” Cronbach’s alpha for this measure = .98.
Sexual Coercion
To measure whether and under what circumstances participants had been coerced into physical sex by their current or most recent romantic partner, participants completed the 34-item SCIRS (Goetz & Shackelford, 2010), adapted only slightly to allow both men and women to report on their partner’s sexually coercive acts (e.g., “My partner pressured me to have sex with him/her against my will.”). The time frame and response scale were the same as described for the SCIRS above. Cronbach’s alpha for this measure = .97.
IPA
Participants completed the 46-item Severity of Violence Against Women Scale (SVAWS; Marshall, 1992), reporting on a 6-point Likert-type scale (0 = never, 5 = a great many times) how often they and their romantic partner had committed each aggressive act (e.g., pushed or shoved, threatened to hurt the other), in the last 12 months of their current or most recent relationship. The SVAWS gives a total score as well as scores for the following subscales: Threats of Violence (19 items), measuring psychologically aggressive threats and intimidation, Acts of Violence (21 items), measuring physical aggression, and Sexual Aggression (six items), measuring forced sexual activity. Responses are summed to create subscale scores; higher scores reflect more aggression. Because we were interested only in participants’ IPA victimization and not their perpetration of IPA, we used only the former scores in our analyses. Cronbach’s alpha for the Full scale and all subscales were as follows: Full scale α = .97, Threats of Violence α = .95, Acts of Violence α = .95, Sexual Aggression α = .89.
Psychological Distress
The Hopkins Symptom Checklist (HSCL; Derogatis, Lipman, Rickels, Uhlenhuth, & Covi, 1974) was used to measure depression and anxiety symptoms. The HSCL is one of the most widely used screening measures of psychological distress in general, and of anxiety and depression in particular. It has been translated and validated in many languages in addition to English, including Arabic, Farsi, Serbo-Croatian, Russian (Kleijn, Hovens, & Rodenburg, 2001). We removed one item (referring to suicidal ideation) because of our inability to follow up with anonymous, online participants who might have expressed suicidal ideation. Participants reported (1 = not at all to 4 = extremely) how much discomfort each symptom (e.g., “feeling fearful” or “feeling lonely”) had caused him or her during the previous 2 weeks (depression α = .93, anxiety α = .88).
Attachment Dysfunction
Participants completed the 12-item short form of the Experiences in Close Relationship Scale (ECR-S; Wei, Russell, Mallinckrodt, & Vogel, 2007). The ECR measures attachment in relationships along two dimensions: attachment anxiety (e.g., “I need a lot of reassurance that I am loved by my partner’’) and attachment avoidance (e.g., “I try to avoid getting too close to my partner’’), via a 1- to 7-point Likert-type scale (1 = disagree strongly, 7 = agree strongly). Previous research has shown that the ECR-S has acceptable test–retest reliability and internal consistency when used with college students (Wei et al., 2007). Cronbach’s alpha for these subscales in this sample were as follows: anxious α = .80 and avoidant α = .82.
Sexual Problems
To assess for sexual problems, we administered the eight-item, Sexual Problems subscale of the Trauma Symptom Checklist (TSC; Briere & Runtz, 1989). The TSC has been validated as a measure of trauma symptoms in non-clinical, undergraduate men and women (Neal & Nagle, 2013). Participants reported, on a 4-point Likert-type scale (1 = never to 4 = often), how often in the last 2 months they experienced various sexual issues, such as “sexual overactivity” and “bad thoughts or feelings during sex” (α = .78).
Results
Similar proportions of men (42%) and women (46%) reported that their current or most recent romantic partner had ever been physically aggressive (SVAWS “acts”), χ2 (1, N = 832) = 1.29, ns, and average frequencies of physical IPA victimization also were similar by gender (Table 1). In contrast, although the percentages of men (58%) and women (62%) who reported psychological victimization were similar, SVAWS “threats”; χ2 (1, N = 834) = 1.11, ns, women experienced more frequent psychologically aggressive threats in their relationships (see Table 1). With regard to sexual aggression and coercion, similar percentages of men (17%) and women (19%) reported that their current or most recent partner had been sexually aggressive toward them: SVAWS “sexual aggression” scale, χ2 (1, N = 819) = 0.59, ns, but proportionally more women (36%) than men (28%) reported experiencing sexual coercion on the SCIRS, χ2 (1, N = 830) = 5.23, p < .05. The average frequency of sexual coercion did not differ by gender (see Table 1). Women experienced more sexting coercion than men both in terms of the average frequency score (Table 1) and in terms of the proportion of women (34%) versus men (25%) who had been victims, χ2 (1, N = 832) = 7.45, p < .01. Correlations between sexual aggression victimization (as measured by the SVAWS Sexual Aggression subscale), sexual coercion victimization, and sexting coercion victimization (both measured via the SCIRS) ranged from .38 to .54 (p < .001 for all) for both men and women. Women reported more UCS (sending a “sext” message despite not really wanting to), symptoms of anxiety and depression, and a younger age at first intercourse; men received significantly more sext messages (see Table 1). Effect sizes for these gender differences were generally small, and men and women scored similarly on most assessed variables (see Table 1).
Descriptive Statistics for Study Variables.
Note. SVAWS = Severity of Violence Against Women Scale, victimization scores; HSCL = Hopkins Symptom Checklist.
p < .05. **p < .01. ***p < .001.
Relations Between Sexting Coercion and Sexual Coercion
Sexting coercion and sexual coercion were highly correlated for both men (r = .8, p < .001) and women (r = .71, p < .001), suggesting that sexting coercion may simply be a part of the sexual coercion construct. However, 19% of all participants had experienced only one type of coercion and not the other. Sexual coercion and sexting coercion, then, do not always go hand in hand. Both, however, relate similarly to distress and dysfunction, as shown in Table 2. In comparing men who had only experienced sexual coercion with those who had only experienced sexting coercion, there were no significant differences in reported levels of psychological distress, sexual problems, or attachment dysfunction. For women, there was a significant difference only for sexual problems, with exclusively sexually coerced women scoring significantly higher than women whose coercion victimization related to sexting alone.
Descriptive Statistics for and Differences Between Those Who Experienced Only Sexual Coercion or Only Sexting Coercion.
Note. HSCL = Hopkins Symptom Checklist.
p < .05.
Coercion Victimization and Sexual Behavior
To assess relations between sexting coercion, sexual coercion, and sexual behavior, we examined correlations between coercion victimization, UCS, UCI, and age at initiation of sexting and intercourse (see Table 3). Again, both forms of coercion were significantly related for both genders, as were age at initiation of intercourse and sexting. Similarly, for men and women, UCS and UCI were positively and significantly correlated. Correlations between coercion victimization and unwanted but consensual sexual behavior differed by gender. For men, sexual coercion and UCI were significantly related. Contrary to predictions, the correlation between sexting coercion victimization and UCS was not significant for men. In contrast, for women, sexting coercion victimization and UCS were much more strongly related than sexual coercion victimization and UCI (see Table 3).
Correlations Between Coercion Victimization, Coerced Behavior, and Age at Initiation of Sexting and Intercourse for Women and Men.
Note. Correlations for women are displayed below the diagonal. Correlations for men are above the diagonal.
p < .05. **p < .01. ***p < .001.
Sexting Coercion, Sexual Coercion, and Polyvictimization
To assess the impact of polyvictimization, we examined cumulative effects of the non-sexual forms of traditional IPA, as measured by the SVAWS (physical and/or psychological aggression), sexual coercion, and sexting coercion. As it has been well established that IPA victims score higher than those without histories of IPA for a number of psychosocial problems, we were not interested in comparing victims with non-victims; only those who experienced IPA were included in these analyses. Participants were placed in one of the following four groups: (a) experienced traditional IPA victimization only, (b) IPA victimization + sexual coercion victimization, (c) IPA victimization + sexting coercion victimization, and (d) IPA victimization + both sexual coercion and sexting coercion victimization. MANCOVAs were run separately for men and women, and were used to compare symptoms of depression and anxiety, sexual problems, and attachment dysfunction across these four groups, controlling for the level of IPA victimization (see Table 4). This was done to ensure that group differences in distress and dysfunction were not simply the result of group differences in IPA victimization, and to compare the impact of coercion victimization over and above the effects of more traditional forms of IPA.
Mental Health Symptoms, Sexual Problems, and Attachment Dysfunction, by IPA Victimization Group (Controlling for Level of IPA Victimization).
Note. Means with differing subscripts within rows are significantly different, after Bonferroni correction. HSCL = Hopkins Symptom Checklist; IPA = intimate partner abuse.
Assumptions for MANCOVA, including homogeneity of variances, were met. The overall MANCOVA was significant for women (Wilk’s λ = .86), F(15,988) = 3.83, p < .0001, η2 = .14, but not for men (Wilk’s λ = .89), F(15,486) = 1.44, p = .12, η2 = .11. However, as suggested by the eta-squared statistic in the men’s model, this appears to be a relatively strong effect which may be obscured by the smaller subsample here. Thus, significant results for both genders will be presented, with more caution warranted in interpreting men’s results. MANCOVA results for men’s data indicated group differences for measures of anxiety, F(4,180) = 5.69, p < .001, η2 = .08; depression, F(4,180) = 3.81, p < .01, η2 = .05; sexual problems, F(4,180) = 6.08, p < .0001, η2 = .05; and avoidant attachment, F(4,180) = 2.71, p < .05, η2 = .002. Results for women suggested group differences for all assessed outcomes: anxiety, F(4,362) = 3.11, p < .05, η2 = .007; depression, F(4,362) = 9.77, p < .0001, η2 = .01; sexual problems, F(4,362) = 14.11, p < .0001, η2 = .04; anxious attachment, F(4,362) = 7.65, p < .0001, η2 = .002; and avoidant attachment, F(4,362) = 7.91, p < .0001, η2 = .02. Bonferroni-corrected, pairwise comparisons are presented in Table 4.
Controlling for differences in the level of IPA victimization across groups, men who experienced IPA plus both forms of coercion reported significantly more sexual problems and symptoms of avoidant attachment than did men who experienced IPA alone, without either form of coercion (see Table 4). Similarly, compared with IPA-only women, women who experienced IPA plus both types of coercions reported significantly more anxiety and more symptoms of avoidant attachment. These women (IPA plus both types of coercions), along with women who experienced IPA plus sexual coercion, also experienced significantly more symptoms of anxious attachment than did IPA-only women. All three groups of coerced women scored higher for depression than IPA-only women. The most complex pattern of group differences involved women’s sexual problems (see Table 4). Women experiencing IPA plus both types of coercions reported the most sexual problems, and significantly more than women in the IPA-only and IPA plus sexting coercion groups. Women who experienced IPA plus sexual coercion also reported significantly more sexual problems than IPA-only women.
Discussion
Abusive, coercive experiences were common in the romantic relationships of these young men and women. Roughly a quarter of the men and a third of women in this sample had experienced sexual coercion, and similar proportions experienced sexting coercion. Even more participants had experienced some physical aggression at the hands of a romantic partner, and more than half had experienced psychologically aggressive threats (e.g., threatening gestures or faces, threats to harm or damage something they cared about) from their current or most recent partner.
Similar frequencies of IPA and sexual coercion victimization were reported by gender, among men and women who had these experiences. More substantial gender differences were found for sexting coercion and UCS. Women reported higher rates and more frequent sexting coercion as compared with men, and engaged in more unwanted sexting. Importantly, gender differences in UCS do not simply reflect gender differences in sexting overall, as men engaged in as much as, or more, sexting as compared with women. The difference, then, is in women’s acquiesce, giving into overt or perceived pressures or feelings of obligation, while men may engage in sexting both more frequently and more willingly. Sexting coercion victimization and unwanted sexting also were highly correlated among women, even higher than the correlation between sexual coercion and unwanted intercourse. These data suggest that women may be even more likely to “give in” to pressures to sext than to have UCI. Perhaps, because of sexual education programming focused on teaching youth to avoid physical sexual behavior or because of informal social messages about sexual rights, girls and young women may feel relatively more empowered or prepared to resist sexual coercion, while they may be less prepared to resist a partner’s coercive attempts to acquire sexually explicit text, picture, or video messages. Specific protocols have been developed to help young women resist coercive sexual advances (e.g., Rowe, Jouriles, McDonald, Platt, & Gomez, 2012). These data suggest that it may benefit young women to have an explicit focus on the topic of sexting coercion, as a component of broader sexual education programs.
For men in this sample, the relation between sexual coercion and unwanted intercourse was much stronger than the relation between sexting coercion and unwanted sexting. Like women, men who experienced the most sexual coercion also reported engaging in unwanted intercourse more often than other participants. Perhaps men’s UCI is mediated by strong feelings of sexual arousal, or perhaps men believe that it is not acceptable or masculine to refuse sex when pressured, as some data on “masculinity scripts” suggest (Stephens & Eaton, 2014). Other research has demonstrated more negative attitudes toward male versus female victims of sexual coercion (Judson, Johnson, & Perez, 2013). Together, data from this and related research suggest that more educational programming may be needed to help dispel myths about gender and sexual coercion (e.g., that men are not victims of sexual coercion) and to teach young men and women alike the skills necessary to resist sexual coercion.
Sexual and sexting coercion victimization were highly correlated for both men and women, but they did not always co-occur. Those who experienced only sexual coercion or only sexting coercion reported, on average, similar levels of psychological distress, attachment dysfunction, and for men, sexual problems. In general, polyvictimization that included some form of coercion was associated with more psychological symptoms, sexual problems, and attachment dysfunction than traditional IPA alone. For women, when IPA was accompanied by a single form of coercion, worse outcomes were generally associated with sexual coercion than sexting coercion. In fact, polyvictimization that included sexual coercion, with or without accompanying sexting coercion, was associated with substantial sexual problems and anxious attachment for women. Sexual coercion, then, may have a relatively stronger impact on women’s sexual health and romantic attachment patterns than sexting coercion, at least in the context of polyvictimization. This supports previous research suggesting that physical and sexual covictimization may represent a distinct form of abuse for women in romantic relationships (Katz et al., 2008).
With few exceptions, a general pattern emerged where traditional IPA alone was associated with the lowest levels of dysfunction, IPA plus one type of coercion was associated with an intermediate level of problems, and IPA plus both types of coercions was associated with the worst outcomes. Thus, although sexual and sexting coercions do not always go hand in hand, when they do co-occur, especially in the context of IPA polyvictimization, they are associated with a particularly high level of distress and dysfunction. Men who experienced all three forms of abuse reported the most psychological distress and more sexual problems and avoidant attachment than men who experienced IPA alone. Women experiencing all three forms of abuse evidenced the most anxiety and attachment dysfunction, scoring higher for both anxious and avoidant attachment than IPA-only women. These findings suggest that sexting coercion is a cumulative risk factor for specific adverse outcomes often associated with more traditional forms of IPA. Furthermore, as suggested by others (e.g., Katz et al., 2008), there appears to be something uniquely distressing about experiencing multiple forms of IPA, as level of traditional IPA victimization (physical and psychological acts) was statistically controlled in this study. In other words, it was not the experience of more severe IPA but of more types of IPA that seemed to predict distress and dysfunction among men and women in this sample.
Sexting coercion, particularly when it leads to acquiesce and engagement in unwanted sexting, is unique from other forms of dating aggression, in that it also has the potential for further psychological trauma due to images being shared with others beyond the intended recipient, or worry that this may happen. This may be of particular concern for women and is potentially evidenced by the finding that women, but not men, report feeling more traumatized by sexting coercion retrospectively—after having time to reflect on the sharing or potential sharing of their “sext” message—than they did at the time the coercion happened (Drouin et al., 2015). However, current definitions of sexual coercion do not include coercion into non-physical acts, for example, sending sexually explicit picture messages. Considering that sexting coercion victimization was related to negative psychological effects and followed the cumulative risk pattern associated with other forms of partner aggression, we suggest that it should be considered a form of digital sexual victimization, and that it warrants further research. As mobile phone usage becomes more normative for younger adolescents and sexting becomes more common, the scope of educational literature addressing unhealthy dating behaviors should be expanded to include coercion related to sexting.
Finally, participants of both genders who began sexting at a young age also initiated intercourse at a young age. Sexting may be a behavioral antecedent to intercourse for some individuals, as suggested by research examining the longitudinal association between sexting and sexual intercourse (Temple & Choi, 2014). There may be some higher-level risk factor that predisposes these youth to engage in both early sex and sexting behaviors (e.g., environmental factors such as social norms in their peer groups or temperamental factors like sensation seeking). Additional data are needed to investigate whether current programs aimed at helping youth to delay the initiation of intercourse might also be effective—or modified so as to be effective—in encouraging youth to delay sexting, as early-onset sexting (in middle school) is associated with sexual activity (Rice et al., 2014).
Limitations and Future Directions
Data in the present study were collected at a single time point, precluding cause-and-effect interpretations. Future research is needed to examine longitudinal relations between polyvictimization that includes sexting coercion and adverse outcomes. In addition, as this work represents an emerging area of inquiry, the present findings should be replicated. Future work in this area could address some of the specific limitations of the present study, for example, by recruiting a more ethnically and geographically diverse sample and by utilizing more extensive measures of the outcomes of interest. Follow-up studies should also examine “unwanted” intercourse and sexting more closely, for example, to tease apart willing but unwanted behaviors (e.g., those performed out of feelings of obligation) from those that are more reflective of active coercion. Data on motives for sexting, particularly motives for unwanted sexting, may be informative here and represent a next step in this line of research. Because of the smaller subsample of men and the fact that some of the male subgroups had relatively few participants, results related to male participants should be considered even more tentative and in need of replication. Some of the null findings for men likely related to low power and some significant group differences may have been obscured.
Despite these limitations, the data suggest that sexting coercion is a fairly common experience for young adults and one that is associated with many of the same negative effects as more traditional forms of IPA. Verbally coerced sexual behavior within a dating relationship is one of the most common forms of sexual victimization for adolescents (Smith et al., 2003). This makes its connection to sexting, a behavior typically associated with teens and college students, an important topic of research as relationships progress in our increasingly digital culture. In addition, those with more psychosocial problems may be entering more unhealthy relationships, where multiple types of abuse might be occurring. Although the direction of influence was not a goal of this study, it is another potential area of exploration in future research. Finally, while we focused on sexting coercion victimization, a reasonable next step is to examine perpetrator characteristics and patterns of sexting coercion perpetration. This line of research could help identify, for example, whether sexting coercion is a fairly common act perpetrated by individuals who are generally unaware of the abusive nature and potential consequences of the behavior, or by a more deviant, manipulative, or abusive subset of individuals.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.