Abstract
Electronic dating aggression among emerging adults is prevalent and has adverse consequences, yet factors that increase or decrease the risk for perpetrating electronic aggression against a romantic partner are understudied. This investigation advances the literature in two novel ways. First, based on intergenerational transmission of violence theories, we tested the link between family-of-origin aggression (FOA) history and electronic aggression toward romantic partners, using a diverse sample of emerging adults. Second, we examined whether perspective-taking and empathy each moderated the association between FOA and electronic dating aggression and explored the moderating role of gender. Participants included 359 undergraduate students (50% female; 42% White) from an urban university. Results indicated that greater FOA during childhood was associated with perpetrating greater electronic aggression against romantic partners. Furthermore, significant interactions indicated that perspective-taking and empathy separately buffered these associations. Youth from aggressive families did not exhibit increased electronic dating perpetration when they had higher perspective-taking or empathy. Males were especially sensitive to the protective effects of perspective-taking. Findings highlight potential points of intervention (i.e., cognitive and affective empathy training) to decrease electronic aggression in romantic relationships and break intergenerational cycles of aggression.
Electronic aggression in emerging adult dating couples is of particular concern due to its prevalence and adverse consequences as an arena for intimate partner violence. Electronic aggression involves using mobile and online platforms to perpetrate hostile (e.g., threats), intrusive (e.g., monitoring whereabouts), and humiliating (e.g., posting an embarrassing picture) behaviors against a romantic partner (Bennett, Guran, Ramos, & Margolin, 2011). Electronic aggression is particularly pernicious due to round-the-clock access afforded by electronic devices and the public and indelible nature of digital footprints. Wide-ranging estimates suggest that up to 70% of college students experience electronic aggression in dating relationships, and electronic aggression has been linked to depression, delinquent behaviors, and concurrent physical and sexual dating victimization (e.g., Reed, Tolman, & Ward, 2016).
Given its pervasiveness, it is important to identify factors that increase or decrease risk for perpetrating electronic dating aggression. In one of the few studies to examine predictors, risky families—characterized by chaos, conflict, neglect, and parental substance use—predicted emerging adults’ electronic aggression perpetration toward peers and dating partners (Kellerman, Margolin, Borofsky, Baucom, & Iturralde, 2013), echoing risk conferred by growing up in aggressive families for engaging in other forms of dating abuse (Ehrensaft et al., 2003). Questions remain, however, regarding whether the intergenerational transmission of family-of-origin aggression (FOA) extends to the electronic domain and what personal characteristics reduce this association. Perspective-taking, or cognitive empathy, involves taking someone else’s point of view and anticipating how that person will interpret a situation; empathy or affective empathy is the capacity to vicariously experience and understand another’s feelings (Davis, 1983). These characteristics may play a particularly crucial role in curbing electronic perpetration because digital interactions lack visual and verbal cues that allow for direct observation of the affective consequences of aggressive behaviors (Runions, Shapka, Dooley, & Modecki, 2013).
The novel contributions of this study are twofold. First, we test the link between FOA history and electronic aggression against romantic partners, using a diverse sample of emerging adults. Second, based on evidence that perspective-taking and empathy are inversely linked with perpetrating relational aggression (Loudin, Loukas, & Robinson, 2003) and adolescent peer cyberbullying (Kowalski, Giumetti, Schroeder, & Lattanner, 2014), we also examine whether these traits reduce the association between FOA and electronic aggression toward a dating partner. We predict that higher FOA is associated with more electronic aggression against dating partners (Hypothesis 1 [H1]); perspective-taking (Hypothesis 2a [H2a]) and empathy (Hypothesis 2b [H2b]) each moderate the association between FOA and electronic perpetration, such that high levels of perspective-taking and empathy break the link between FOA and electronic aggression. We also explore the role of gender in all study hypotheses, given mixed findings regarding gender differences in the amount of electronic aggression perpetration committed, its target, and its emotional effect (e.g., Bennett et al., 2011; Reed, Tolman, & Ward, 2017).
Method
Participants and Procedure
Participants were 359 undergraduate students (181 female) aged 18 to 27 years (M = 20.34, SD = 1.54) from an urban university, who anonymously completed an online survey for extra credit. Inclusion required participants to be English-speaking and to have completed at least 70% of the items. Self-reported race was 0.8% American Indian/Alaskan Native, 32.6% Asian, 5.6% African American/Black, 42.3% Caucasian, 7.0% multiracial, 5.0% other, and 6.7% declined to report; 15.0% identified as Hispanic/Latino. The university institutional review board (IRB) approved study procedures.
Measures
Electronic aggression perpetration
Past year electronic aggression against a dating partner was assessed using four items adapted from the How Friends Treat Each Other Questionnaire (Bennett et al., 2011). One item tapped hostile behaviors (Sent a mean, hurtful, or threatening email or text message); one measured humiliation (Posted something mean, hurtful, or insulting on a social networking site); and two assessed intrusive, cyber-stalking behaviors (e.g., Intrusively called or texted to monitor or check up on her or him). For each item, participants rated how often they engaged in the behavior within the past year from 0 (never happened) to 4 (more than 10 times). Mean scores were calculated across the four items (α = .83, M = 0.13, SD = 0.40).
FOA
Using a modified Domestic Conflict Inventory (Margolin, John, & Foo, 1998) and Conflict Tactics Scales–Parent/Child (Straus, Hamby, Finkelhor, Moore, & Runyan, 1998), we assessed two types of FOA—(a) parent-to-child aggression (10 items, for example, Swore or cursed at you; Hit you with a hand or object) and (b) parent-to-parent aggression (same 11 items repeating for father and mother, for example, Insulted or shamed other parent in front of others; Shook or slapped other parent)—using a scale from 0 (none) to 4 (>6 times). We averaged the scores across all items to create a total FOA score (α = .94, M = 0.70, SD = 0.80).
Perspective-taking and empathic concern
We used the Perspective-Taking and Empathic Concern scales from the Interpersonal Reactivity Index (Davis, 1983). The Perspective-Taking scale, scored as the sum of seven items (α = .77, M = 11.38, SD = 3.10), assesses the ability to understand the viewpoint of others. The Empathic Concern scale, also the sum of seven items (α = .75, M = 13.08, SD = 2.69), measures feelings of compassion for others. Responses ranged from 0 (does not describe me well) to 3 (does describe me well).
Analytic Plan
We tested for differences in key study variables by race/ethnicity (dichotomously coded as White/non-White, Black/non-Black, Asian/non-Asian, Hispanic/non-Hispanic) and socioeconomic status (SES), as indexed by parent’s highest level of education. Because only race/ethnicity variables altered the significance and magnitude of the predictor, we adjusted for race and ethnicity as covariates in all hypothesis testing and excluded SES to increase parsimony. Study hypotheses were tested using hierarchical linear regression. Covariates were entered at Step 1, predictors entered at Step 2, two-way interactions entered at Step 3, and three-way gender interactions entered at Step 4. Slopes were calculated for significant interactions and plotted at high (+1 SD) and low (–1 SD) values of continuous moderators. Regions of significance (ROS) were calculated using the Johnson–Neyman technique to test the limits of observed effects.
Results
In the past year, 16.3% of participants reported perpetrating at least one act of electronic aggression against a dating partner. Mean scores did not significantly differ by gender for electronic aggression perpetration, perspective-taking, or FOA, though females reported greater empathy, t(357) = 4.55, p < .001. Across all participants, FOA was associated with more electronic perpetration (r = .36, p < .001), and the two proposed moderators (perspective-taking and empathy) were moderately associated (r = .44, p < .001). Both greater perspective-taking (r = –.14, p < .01) and greater empathy (r = –.12, p < .05) were associated with reduced electronic aggression.
In support of H1, participants who experienced greater FOA reported greater electronic aggression (B = .16, SE = .02, p < .001); this effect did not differ by gender (B = –.09, SE = .05, p = .08). As displayed in Table 1, in support of H2a, perspective-taking moderated the association between FOA and electronic aggression (B = –.05, SE = .01, p < .001), and this effect differed by gender (B = .03, SE = .02, p = .04) such that the buffering effect of high perspective-taking was more pronounced for males (see Figure 1a). ROS show that the association between FOA and electronic aggression becomes nonsignificant at perspective-taking values of 0.46 SD above the mean for males and 0.93 SD for females. In support of H2b, empathy also moderated the link between FOA and electronic aggression (B = –.04, SE = .01, p < .001), but this effect did not differ by gender. As shown in Figure 1b, for both males and females, at low levels of empathy, FOA was associated with greater electronic perpetration, but when empathy was high, the association between FOA and electronic aggression was nonsignificant. The link between FOA and electronic perpetration became nonsignificant at empathy levels of 0.66 SD.
Regression Analyses Examining Perspective-Taking and Empathy as Moderators between FOA and Electronic Aggression Perpetration Against Dating Partner.
Note. All values from final step. Two moderators are tested: perspective-taking and empathy. FOA = family-of-origin aggression.
p < .05. **p < .01. ***p < .001.
Plot of interactions between FOA and (a) perspective-taking, (b) empathy on electronic aggression perpetration against dating partner.
Discussion
The present study extends the intergenerational spillover of aggression to the electronic domain and also identifies moderators of this association. As predicted, FOA was associated with increased electronic perpetration toward a romantic partner (H1). To our knowledge, this is the first study to link family aggression with electronic dating aggression. Findings suggest that those exposed to aggressive family interactions during childhood may be more likely to use aggression as a strategy for dealing with romantic conflicts whether online or in-person. Results are consistent with conflict sensitization theories (e.g., Fite et al., 2008), such that FOA-exposed youth may interpret ambiguous electronic communications from partners as hostile and respond aggressively.
Still, perspective-taking and empathy each attenuated the link between FOA and electronic aggression (H2). Whereas empathy was similarly protective for both males and females, the buffering effect of perspective-taking was more pronounced for males, with even mild levels mitigating the effects of FOA. These findings are consistent with previous research demonstrating that adolescent males are more sensitive to the protective effects of perspective-taking compared with empathy (e.g., Ang & Goh, 2010). Moreover, given prior findings that females find electronic aggression more distressing than do males (e.g., Reed et al., 2017), perspective-taking may prompt males to consider whether engaging in electronic aggression would upset their partners.
The ability to take others’ viewpoints or perceive their emotions may be especially important in electronic communication, as messages lack important visual and verbal cues and can occur in quick-fire succession (Runions et al., 2013). Although perspective-taking and empathy are only modestly associated, it is difficult to parse their separate influences in attenuating electronic aggression. However, particularly in the context of ambiguous messages, those with high perspective-taking and/or empathy may be able to consider alternative meanings and motivations behind electronic communications, to accept more benign interpretations, to tune in to the partner’s potential negative reactions, and to use these reflective capacities to regulate their own emotions.
Our findings highlight potential points of intervention for youth, such as skills building in perspective-taking and empathy to break intergenerational cycles of aggression. Programs designed to bolster cognitive and affective empathy to reduce anger and in-person aggression (e.g., Reddy et al., 2013) might also be effective in decreasing aggression in the electronic domain.
Results should be interpreted in light of several limitations. First, we relied on retrospective, self-report data, which introduce recall bias and shared method variance. Future studies could use prospective, multiple-informant measures to address underreporting of aggressive events and to capture the direction of effects across time. Second, our rate of electronic dating aggression was somewhat lower than other studies of college students (e.g., Kellerman et al., 2013; Reed et al., 2016) which may be related to our use of an abbreviated four-item measure; future research might benefit from a more detailed assessment of electronic aggression. Third, alternative methods (e.g., downloading text message logs and using ecological momentary assessment) would enable the assessment of electronic aggression in real time and identify precipitants and reactions. Fourth, our sample was constrained to college students, limiting generalizability, as emerging adults who attend college may exhibit resiliencies relative to those who do not attend college. Finally, despite our sample’s ethnic and racial diversity, the sample size did not permit meaningful exploration of diversity (beyond gender) as a moderator of these important links in intergenerational transmission phenomena.
Despite these limitations, this study expands the intergenerational transmission of aggression to the electronic domain and highlights individual characteristics that may protect emerging adults against this continuity of aggression. In doing so, we identify factors to inform intervention efforts for dating violence in the digital age: Perspective-taking and empathy-building skills may be critical components for university health promotion programs aimed at minimizing electronic dating aggression overall, and with specific impacts for those exposed to FOA.
Footnotes
Acknowledgements
The authors extend their appreciation to USC Family Studies Project colleagues, especially Hannah Rasmussen, Corey Petit, Kristine Hossepian, Vivian Rotenstein, Reout Arbel, and Diana Bennett for their support in this project, including their feedback on previous drafts of this article.
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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project is based on work supported by NIH-NICHD R21-HD072170 A1 (Margolin, Principal investigator [PI]), NSF BCS-1627272 (Margolin, PI), and NSF GRFP DGE-0937362 (Miller, PI).