Teen perceptions of adolescent dating violence
Abstract
Introduction
Previous research shows that adolescents who experience dating violence most often disclose their victimization to a peer or friend, more so than to other sources of support. However, surprisingly little research has explored how adolescents respond to peer disclosures of dating violence. Addressing this gap, the present study assessed variations in adolescents' perceptions of blame, interpretations of the incident as violence, and intentions to respond across physical, psychological, sexual, cyber-psychological, and cyber-sexual dating violence scenarios.
Methods
As part of a national research project across Canada, 663 high school adolescents (432 girls, 65.2%) between the ages of 14-17 were randomly assigned to complete a questionnaire which included one of five different hypothetical dating violence scenarios. Next, participants responded to questions about their perceptions of the incident, as well as victim and perpetrator blame and responsibility, and their intentions to respond.
Results
Results indicated that the type of dating violence experienced and the age and gender of participants all played a role in perceptions of blame, understandings of violence, and intentions to respond.
Conclusions
As one of the first studies to explore how adolescents perceived and responded to dating violence, considering both in-person and cyber forms of dating violence, this study fills an important gap in the literature. Findings underscore the uniqueness of cyber forms of dating violence and how pre/intervention programs must address the specific contexts and issues unique to each type of dating violence.
1 INTRODUCTION
It is estimated that about one-third of teens will experience some form of dating violence during their adolescence (Exner-Cortens et al., 2021; Leen et al., 2013). Of these youth, approximately 43% will seek help from and disclose their victimization to a friend or peer, more than to any other source of support, including school staff and family members (Bundock et al., 2020). Though adolescents most often reach out to peers for support, little is known about how teens perceive and react to disclosures of dating violence. Understanding peer reactions is crucial, as both adolescent and adult victims of dating violence report that receiving a response that is unhelpful or blames them for their victimization significantly impacts overall well-being, recovery, and whether they choose to stay in their abusive relationship (Jackson et al., 2000; Moe, 2007). Extending prior adult research, this study explored how adolescents respond to peer disclosures of dating violence. Given their more limited world experiences, and their still-developing cognitive functioning and empathic development, teens may respond differently than adults to peer disclosures of dating violence. Using hypothetical scenarios, we investigated how perceptions of responsibility and blame, interpretations of the incident as violence, and intentions to respond differed if a peer disclosed an experience of physical, sexual, psychological, cyber-psychological, or cyber-sexual dating violence.
Dating violence is defined as an intentional act or threat of violence by a dating partner or ex-partner, and generally takes the form of physical, psychological, and/or sexual violence (Breiding et al., 2015). Following Breiding et al., physical dating violence involves an intentional physical act to harm one's partner, or threats of physical harm (e.g., slapping, hitting, using a weapon, physically restraining or preventing a partner from leaving). Psychological dating violence includes using words or actions to control one's partner and/or harm their emotional or mental well-being (e.g., insults, yelling, controlling or monitoring a partner's whereabouts or with whom they interact, using emotional manipulation to engender guilt). Sexual dating violence refers to actual or attempted nonconsensual bodily contact for a sexual purpose, (e.g., unwanted kissing/touching, threatening or manipulating a partner into sex, rape, or attempted rape).
Beyond in-person violence, the last 15 years have seen the emergence of cyber forms of dating violence which involve “the control, harassment, stalking, and abuse of one's dating partner via technology and social media” (Zweig et al., 2014, p. 1306). This includes both cyber-psychological dating violence, defined as using technology to access a partner's social media/text messages or monitor a partner's activities, and cyber-sexual dating violence, which involves sending sexually explicit photos or text messages without consent (Draucker & Martsolf, 2010; Zweig et al., 2013). One factor that distinguishes cyber dating violence from in-person forms is that perpetrators can continuously contact, monitor, and abuse their victims through phone or social media (Draucker & Martsolf, 2010). Thus, such violence may be longer lasting, as perpetrators can repeatedly text, call, or email their victims, or post humiliating or private photos and content on various online platforms (Draucker & Martsolf, 2010). Given its recent emergence, it is unclear whether cyber dating violence is a distinct form of dating violence, or a new medium through which traditional forms of dating violence can be perpetrated.
2 LITERATURE REVIEW1
2.1 Adolescent attitudes toward dating violence
Overall, adolescents are not very accepting of dating violence (Price et al., 1999; Reeves & Orpinas, 2012), although teens view some violent actions as more tolerable or justified in dating relationships, particularly when harm is not obvious, or when “joking around” (e.g., Sears et al., 2006). Adolescents report greater acceptance of psychological dating violence, followed by physical dating violence, and are least accepting of sexual dating violence (Price et al., 1999). Considering participant characteristics, adolescent boys are more accepting of dating violence than girls, and younger adolescents are more tolerant and accepting of dating violence than older students (Courtain & Glowacz, 2021; Price et al., 1999). However, research on perceived acceptability offers only a limited picture of the problem. When explicitly asked, most youth do not condone dating violence (Price et al., 1999; Reeves & Orpinas, 2012). However, a different picture emerges in adult research that assesses perceptions of victims and perpetrators of dating violence, offering a more nuanced understanding of the issue.
2.2 Victim blaming
Victim blaming occurs when a victim is viewed as responsible in some way for their assault, and the perpetrator is exonerated of responsibility to some degree. Beyond explicit blame and responsibility, victim blaming research also considers perceptions of severity, amount of trauma experienced, event ambiguity, intent to harm, and sympathy toward both victim and perpetrator (Davies et al., 2006; Rogers & Davies, 2007). Research with adults shows that victim blaming can significantly impact a victim's well-being and recovery, with negative reactions or a lack of support after an assault linked to poor health outcomes in victims, including posttraumatic stress, depression, and self-blame (e.g., Moe, 2007). Given the limited studies of victim blaming among adolescents, adult research provides a useful backdrop for the current study.
Adult men are consistently found to be more accepting of dating violence, are less likely to label dating violence as abuse, and assign less blame to the perpetrator, compared to women, although sometimes this difference is modest (e.g., Capezza & Arriaga, 2008; Dardis et al., 2017). In contrast, women attribute less responsibility to victims of dating violence and view the perpetrator's actions more negatively than men (Capezza & Arriaga, 2008; Sylaska & Walters, 2014). Finally, in a study of responses to a physically violent domestic dispute, Seelau and Seelau (2005) found that women were more likely to use systemic interventions (e.g., contacting police), whereas men were more likely to try to talk to the couple or do nothing.
Considering violence type, adults view physical and sexual dating violence as more severe and abusive than psychological violence (Capezza & Arriaga, 2008; Dardis et al., 2017), and ascribe less fault to victims of physical or sexual dating violence than to victims of psychological violence (Taylor & Sorenson, 2005) Despite such perceptions, victims of psychological violence also face many negative consequences that adversely impact their well-being (e.g., Foshee et al., 2013). Of interest in this study was whether similar perceptions are evident among adolescents.
Few studies have investigated victim blaming in the context of adolescent dating violence, and even here the focus has been on adult perceptions. For example, Taylor and Sorenson (2004) assessed adult reactions to vignettes depicting physical, sexual, or psychological dating violence within boy-girl adolescent relationships, asking whether the violence depicted was right or wrong, illegal, and in need of police intervention. Although 97% of adults viewed all types of dating violence as wrong, they were more likely to view physical and sexual violence (and threats of such violence) as illegal and in need of police or legal intervention. Adults also tended to favor legal/police intervention more as severity increased, with severe sexual violence (i.e., rape) seen as most in need of legal sanctions. Victim blame research, however, extends beyond questions of right/wrong, legality, or need for intervention. Qualitative studies show that, even when dating violence is viewed as wrong, people may still tolerate, accept, or justify such violence, and may blame the victim for “causing” the violence or bringing it on themselves (e.g., Bowen et al., 2013).
2.3 Present study
The current study extends previous research by exploring perceptions of dating violence among adolescents, assessing perceptions and reactions to disclosures of both traditional, in-person and cyber forms of dating violence. Using a between-subjects design, adolescents were asked to respond to one of five hypothetical vignettes depicting physical, sexual, psychological, cyber-psychological, or cyber-sexual dating violence, as disclosed to them by a peer. Perceptions of blame and responsibility for both victim and perpetrator, their understandings of the situation as violence, and their likelihood of responding were assessed. Replicating previous adult research, we considered the three hypotheses. Specifically, it was expected that (1) older adolescents would view all dating violence situations as more violent than younger adolescents, (2) adolescent boys would assign more blame and responsibility to victims than perpetrators, relative to girls, and (3) physical and sexual acts would be considered more violent than psychological acts, for both face-to-face and cyber forms. Of particular interest was whether cyber forms of dating violence were viewed differently than their traditionally recognized, in-person counterparts. Although dating violence can occur within all dating relationships, regardless of age, gender, or sexual orientation, this study focused solely on girl victims/boy perpetrators within heterosexual relationships.
3 METHODOLOGY
Data were collected as part of a larger, national project on teen dating violence, supported by the Public Health Agency of Canada, and conducted by researchers within PREVNet, a national organization of researchers and child-serving agencies across Canada with the aim of Promoting Relationships and Eliminating Violence in children and youth. Data were collected from high school students in Ontario, Quebec, and Manitoba, Canada in late 2019 and early 2020. Only measures used in the current study are described below.
3.1 Participants
From an initial sample of 723 high school students, two participants were removed for invalid responding, and 42 were outside the specified age range (14–17 years). Another 16 students who identified as gender-expansive or nonbinary were omitted due to insufficient participant numbers for analyses beyond the consideration of boys and girls. The final sample included 663 adolescents (432 girls, 231 boys) that were divided into two groups, with 53% (n = 350) aged 14–15 and 47% (n = 313) aged 16–17. Most participants (81%) selfidentified as heterosexual, 8% as bisexual, 3% as gay or lesbian2, 4% as questioning, 3% as “something else”; and 1% preferred not to answer. Regarding ethnicity, 42% selfidentified as White/European, 11% as Southeast Asian, 10% as Mixed Race, 6% as African/Caribbean, 6% as South Asian, 4% as First Nations, 3% as East Asian, 5% as “something else,” and 13% not answering or unsure. Procedure. The survey was completed entirely online using survey software (Qualtrics) during a 30-40-min classroom period in participating schools. Students who received parental consent for the study, and who themselves provided assent, were informed that their participation was voluntary and that they could withdraw from the study at any time without penalty.
3.2 Measures
3.2.1 Demographic information
Participants were asked to provide information on their age, gender, ethnicity, current relationship status, and sexual orientation, as described above.
3.2.2 Dating violence vignettes
Participants were randomly assigned to one of five vignette conditions (see Appendix A), and only read one vignette. Each vignette depicted a hypothetical dating violence scenario between two students, though the term “dating violence” was not explicitly mentioned. The vignettes were comparable in length (about 300 words) and written as similarly as possible, with an identical introduction, context, and conclusion, varying only in the dating violence depicted (physical, psychological, sexual, cyber-psychological, or cyber-sexual). The victim was always a girl (“Anna”) and the perpetrator was always a boy (“Liam”), both in the same grade as the participant. In each scenario, Anna comes to speak with the participant, a peer she has come to know over the past year, and informs them of a “confusing” experience she had with her boyfriend, Liam. Unsure about what to do, Anna asks the participant what they think about the situation.
3.2.3 Blame questionnaire
A 10-item blame questionnaire, adapted from Davies et al. (2006), assessed participants' attributions of blame and responsibility toward both victim and perpetrator. Responses to all items were made on 7-point, Likert scales, ranging from 1 (pro-victim) to 7 (anti-victim). The Reaction toward the Victim subscale assessed participants' perceptions of the violence in terms of severity, victim responsibility, level of trauma, and negative impact likely experienced by the victim (e.g., “How responsible do you think Anna was for what happened to her in the band room?” “How much do you think Anna actually enjoyed the attention she received from Liam?”). The Reaction toward the Perpetrator subscale assessed perpetrator responsibility, possible motivations, and whether the perpetrator should receive any punishment (e.g., “Should Liam be punished for what he did to Anna?” “Liam did not mean to hurt Anna, he just got carried away. How much do you agree?”).
Although both scales have shown adequate internal consistency in past research (e.g., Davies et al., 2006), internal consistency in this sample was low (α = .548 for the victim subscale, α = .599 for the perpetrator subscale), with low to moderate intercorrelations among the 10 items, 80% of which were under the 0.3 level (see Table 1). A principal component analysis (PCA) yielded three factors with eigenvalues >1.00, accounting 28%, 20%, and 11% of the total variance. The first and largest factor, however, was uninterpretable. These findings suggest that the 10 items of the blame scale were relatively independent in the present adolescent sample. Accordingly, following Huberty and Morris (1989), these items were more appropriately evaluated separately. An additional item was included to assess whether the situation was viewed as dating violence (i.e., “Liam committed an act of dating abuse/violence. How much do you agree with this statement?”).
| BI2 | BI3a | BI4a | BI5 | BI6a | BI7a | BI8 | BI9 | BI10 | |
|---|---|---|---|---|---|---|---|---|---|
| BI1a School Admin Seriousness | −.179** | .182** | .440** | −.051 | .180** | .195** | −.024 | .066 | −.074* |
| BI2 Anna responsibility | .222** | −.018 | .309** | −.010 | .002 | .164** | .150** | .455** | |
| BI3a Liam responsibility | .272** | .133** | .154** | .186** | .152** | .207** | .280** | ||
| BI4a Liam punishment | .022 | .329** | .306** | .179** | .315** | .099** | |||
| BI5 Anna enjoyment | .184** | .072* | .233** | .202** | .411** | ||||
| BI6a Anna upset | .354** | .090* | .170** | .132** | |||||
| BI7a Anna negative impact | .169** | .268** | .091* | ||||||
| BI8 Liam behavior | .489** | .270** | |||||||
| BI9 Liam intentions | .260** |
- Note: BI denotes blame item, a denotes reversed scored. All correlations are one–tailed. BI10 along the top row is the Anna Behavior Blame item. The 11th blame item on whether participants thought Liam committed an act of dating violence was an addition to the scale by the authors, thus was not included in the PCA and correlation analyses.
- * p < .05 level
- ** p < .001 level.
3.2.4 Responses to dating violence scenario
Participants were asked how likely they would be to engage in each of 17 actions in response to the dating violence disclosed (e.g., do nothing, talk to Anna's parent[s]/guardian, encourage Anna to break up, reassure Anna that she is not to blame). Potential responses were derived from previous research (Casey et al., 2018; Futerman, 2014; Sylaska & Walters, 2014), and were expanded for the current study. Prior dating violence research has focused on formal supports (e.g., police), with less attention given to other sources of social support or more informal responses, such as listening to the victim (Sylaska & Walters, 2014). Both formal and informal responses were considered in the present study.
Intercorrelations among the 17 items (see Table 2) ranged from 0.01 to 0.77. A PCA was conducted to consider possible data reduction. As shown in Table 2, three items that did not demonstrate any correlations greater than 0.3 (do nothing, talk to Liam on Anna's behalf, let Anna sort it out with Liam) were dropped from the PCA, but considered individually in analyses. Results of the PCA (varimax rotation) conducted with the remaining 14 items yielded three components with eigenvalues greater than one, explaining 33%, 17%, and 9% of the total variance, respectively. Factor one included five items reflecting efforts to reach out to various adults (notify the principal, notify a teacher, notify a counselor, contact the police, and talk to Anna's parent/guardian), with high internal consistency (α = .880). Factor two included six items related to supporting or encouraging the victim (reassure Anna that she is not to blame, help Anna decide on what to do next, listen to Anna, encourage Anna to speak with a trusted adult, encourage Anna to talk to her parent/guardian, encourage Anna to break-up with Liam) and showed acceptable internal consistency (α = .746). Factor three included three items regarding reaching out to peers and/or your own parents (talk to friends and ask them what to do, encourage Anna to talk to her friends, and talk to your own parent/guardian about what you should do), but demonstrated low internal consistency (α = .552). Given the low internal consistency and our interest in the items in this third factor, each was evaluated individually in subsequent analyses.
| 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 Do nothing | −.139** | −.174** | −.174** | −.116** | −.134** | −.211** | .238** | −.115** | −.006 | −.135** | −.244** | −.228** | −.138** | −.190** | −.104** | −.133** |
| 2 Talk to A's parent | .223** | .378** | .042 | .172** | .333** | −.135** | .419** | .154** | −.006 | .042 | .086* | .501** | .437** | .525** | .469** | |
| 3 Talk to Liam | .009 | .164** | .077* | .084* | −.034 | .045 | .154** | .037 | .080* | .109** | .092* | .103** | .110** | .110** | ||
| 4 Enc. A to talk to parents | .280** | .309** | .636** | −.160** | .405** | .154** | .238** | .326** | .268** | .335** | .302** | .333** | .283** | |||
| 5 Enc. A to talk to friend | .211** | .264** | .029 | .126** | .379** | .237** | .278** | .247** | .039 | .094** | −.001 | −.040 | ||||
| 6 Enc. A to breakup w/Liam | .310** | −.239** | .138** | .137** | .093* | .289** | .306** | .135** | .150** | .189** | .223** | |||||
| 7 Enc. A to speak to adult | −.194** | .323** | .122** | .294** | .363** | .377** | .351** | .396** | .352** | .243** | ||||||
| 8 Let A sort out w/Liam | −.123** | .016 | −.070* | −.188** | −.191** | −.171** | −.194** | −.166** | −.203** | |||||||
| 9 Talk to own parent | .363** | .087* | .122** | .139** | .386** | .308** | .354** | .256** | ||||||||
| 10 Talk to own friend | .136** | .128** | .105** | .200** | .158** | .160** | .083* | |||||||||
| 11 Listen to A | .479** | .367** | .015 | .066* | .013 | −.007 | ||||||||||
| 12 Reassure A not to blame | .490** | .094** | .131** | .076* | .015 | |||||||||||
| 13 Help A make decision | .140** | .156** | .094** | .040 | ||||||||||||
| 14 Notify teacher | .710** | .769** | .531** | |||||||||||||
| 15 Notify counselor | .707** | .517** | ||||||||||||||
| 16 Notify principal | .689** |
- Note: “Enc.” denotes encourage; “A” denotes Anna. Item 17 along the top row is the Notify Police item. All correlations are one–tailed.
- * p < .05 level
- ** p < .01 level.
4 RESULTS
4.1 Analyses for perceptions of blame items
A series of between subjects, two (gender) by two (age group) by five (type of dating violence) analyses of variance (ANOVAs) were conducted to evaluate perceptions of blame and violence, one for each of the 11 items. Given the number of analyses conducted for the blame and violence items, an adjusted alpha level of 0.0045 was used to control for type I error in determining significance. Across analyses, all two- and three-way interactions between age, gender, and vignette condition were nonsignificant, and thus are not reported. Main effects for gender and age are presented first, followed by main effects of vignette condition, for which follow-up, pairwise analyses (Bonferroni) were conducted to detect significant differences across vignette conditions.
4.1.1 Gender main effects
Results indicated significant main effects for gender in 6 of the 11 analyses conducted. Means and standard deviations for each item across gender and age groups are presented in Table 3. As shown in Table 3, girls were slightly more likely to rate the incidents as dating violence than boys (marginal effect), with no gender differences in perceptions of how seriously school administrators would consider the incident. Nonsignificant gender differences were also observed for students' ratings of how upset the victim would be, how responsible the perpetrator was, or in whether the perpetrator should be punished. However, relative to girls, boys attributed more responsibility and blame to victims, and were more likely to suggest that the victim enjoyed the attention she received. Boys were more likely to attribute the perpetrator's behavior to liking the victim and were more likely to suggest that the perpetrator just got carried away. boys also tended to rate the event as less likely to impact the victim's future relationships (marginal effect).
| Items (1–7 response scale) | Gender differences | Age differences | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| F(df) | p | partial η2 | Boys M (SD) | Girls M (SD) | F(df) | p | Partial η2 | Younger M (SD) | Older M (SD) | |
BI11: Liam committed act of dating violence? (strongly disagree–strongly agree) |
F(1, 620) = 6.70 | .010† | .01 | 5.63 (1.63) | 5.94 (1.37) | F(1, 629) = 0.282 | .596 | <.01 | 5.75 (1.50) | 5.82 (1.43) |
BI1: If reported, how seriously would school administrators take it? (not seriously–very seriously) |
F(1, 629) = 4.77 | .029 | .01 | 5.04 (1.69) | 4.77 (1.60) | F(1, 629) = 4.29 | .039 | .01 | 5.03 (1.63) | 4.77 (1.63) |
BI2: How responsible was Anna? (not responsible–very responsible) |
F(1,625) = 20.76 | <.001* | .03 | 2.94 (1.87) | 2.33 (1.55) | F(1,625) = 1.82 | .178 | <.01 | 2.73 (1.70) | 2.54 (1.67) |
BI10: How much was Anna's behavior to blame? (no blame –high blame) |
F(1,621) = 17.80 | <.001* | .03 | 2.48 (1.57) | 2.01 (1.23) | F(1,621) = 11.56 | <.001* | .02 | 2.43 (1.45) | 2.05 (1.25) |
BI6: How upset is Anna by this event? (not upset–very upset) |
F(1,627) = 2.58 | .109 | <.01 | 5.74 (1.36) | 5.91 (1.24) | F(1,627) = 1.82 | .177 | <.01 | 5.90 (1.25) | 5.76 (1.33) |
BI7: How likely would event impact Anna's future romantic relationships? (not likely–very likely) |
F(1,626) = 7.74 | .006† | .01 | 5.29 (1.36) | 5.58 (1.22) | F(1,626) = 0.07 | .789 | <.01 | 5.42 (1.24) | 5.45 (1.31) |
BI5: How much did Anna enjoy the attention? (no pleasure–high pleasure) |
F(1,621) = 10.53 | .001* | .02 | 2.43 (1.63) | 2.02 (1.38) | F(1,621) = 1.69 | .194 | <.01 | 2.31 (1.49) | 2.14 (1.48) |
BI3: To what extent is Liam responsible? (not responsible–very responsible) |
F(1,626) = 3.89 | .049 | .01 | 6.23 (1.32) | 6.44 (1.21) | F(1,626) = 1.42 | .234 | <.01 | 6.27 (1.35) | 6.40 (1.13) |
BI4: Should Liam be punished? (no punishment–severe punishment) |
F(1,624) = 5.79 | .016 | .01 | 5.41 (1.58) | 5.67 (1.31) | F(1,624) = 2.62 | .106 | <.01 | 5.63 (1.39) | 5.45 (1.44) |
BI8: Liam would not have behaved this way unless he really liked victim (strongly disagree–strongly agree) |
F(1,614) = 45.27 | <.001* | .07 | 3.92 (1.85) | 2.90 (1.78) | F(1,614) = 9.01 | .003* | .01 | 3.64 (1.91) | 3.18 (1.80) |
BI9: Liam just got carried away? (strongly disagree–strongly agree) |
F(1,619) = 47.85 | <.001* | .07 | 3.52 (1.84) | 2.58 (1.48) | F(1,619) = 1.08 | .299 | <.01 | 3.12 (1.70) | 2.98 (1.64) |
- Note: Significant mean differences at p < .005 are indicated with *. Marginally significant trends at p = .005−.01 are indicated with†.
4.1.2 Age main effects
Results indicated significant main effects of age for only 2 of the 11 analyses conducted (see Table 3). Specifically, younger students (age 14–15) were more likely than their older peers (age 16–17) to blame victims for the incident and to suggest that the perpetrator would not have behaved this way unless he really liked the victim.
4.1.3 Differences across types of dating violence
Results indicated significant differences in perceptions across dating violence conditions for 9 of the 11 analyses conducted. Means and standard deviations for each item are presented in Table 4.
| Items | Type of dating violence | Significant post-hoc effects | ||||
|---|---|---|---|---|---|---|
| PHYS | PSYC | SEX | C-PSYC | C-SEX | ||
| M (SD) | M (SD) | M (SD) | M (SD) | M (SD) | ||
| (n = 131–138) | (n = 125–130) | (n = 114–119) | (n = 145–147) | (n = 116–118) | ||
Perpetrator committed act of dating violence? (strongly disagree–strongly agree) |
6.24 (1.24) | 5.37 (1.63) | 6.24 (1.14) | 5.18 (1.50) | 5.90 (1.37) | Phys, Sex > Psyc, C-P C-S > C-P |
If reported, how seriously would school administrators take it? (not seriously–very seriously) |
4.94 (1.56) | 4.23 (1.52) | 5.67 (1.33) | 4.07 (1.58) | 5.61 (1.46) | Sex, C-S > Phys, Psyc, C-P Phys > Psyc, C-P |
How responsible was victim? (not responsible–very responsible) |
1.98 (1.51) | 2.23 (1.51) | 2.66 (1.67) | 2.65 (1.54) | 3.66 (1.84) | C-S > Phys, Psyc, Sex, C-P Sex, C-P > Phys |
How much was victim's behavior to blame? (no blame–high blame) |
1.75 (1.16) | 2.01 (1.38) | 2.31 (1.38) | 2.77 (1.44) | 2.38 (1.32) | C-S > Phys, Psyc Sex, C-P > Phys |
How upset is victim by this event? (not upset–very upset) |
5.89 (1.27) | 5.80 (1.31) | 5.75 (1.28) | 6.36 (0.93) | 5.33 (1.31) | Phys > C-P C-S > Psyc, C-P |
How likely would event impact victim's future romantic relationships? (not likely–very likely) |
5.35 (1.27) | 5.06 (1.36) | 5.65 (1.26) | 5.24 (1.25) | 5.89 (1.02) | C-S > Phys, Psyc, C-P Sex > Psyc |
How much did victim enjoy the attention? (no pleasure–high pleasure) |
2.23 (1.52) | 1.89 (1.35) | 2.44 (1.46) | 2.26 (1.45) | 2.30 (1.60) | All nonsignificant |
To what extent is perpetrator responsible? (not responsible–very responsible) |
6.48 (1.26) | 6.20 (1.28) | 6.47 (1.04) | 6.13 (1.47) | 6.40 (1.11) | All nonsignificant |
Should perpetrator be punished? (no punishment–severe punishment) |
5.80 (1.40) | 4.79 (1.54) | 6.13 (1.03) | 4.83 (1.33) | 6.15 (0.99) | Phys, Sex, C-S > Psyc, C-P |
Perpetrator would not have behaved this way unless he really liked victim? (strongly disagree–strongly agree) |
3.31 (1.88) | 3.66 (1.83) | 3.52 (1.94) | 3.71 (1.81) | 2.84 (1.86) | Psyc, C-P > C-S |
Perpetrator just got carried away? (strongly disagree–strongly agree) |
2.75 (1.67) | 3.17 (1.61) | 3.22 (1.81) | 3.42 (1.55) | 2.68 (1.63) | C-P > Phys, C-S |
- Abbreviations: C-P, cyber-psychological; C-S, cyber-sexual; PHYS, physical; PSYC, psychological; SEX, sexual. Significant differences are bolded.
Although students generally considered all scenarios to describe acts of dating violence (all means > 5 out of 7), significant condition effects were observed, F(4, 620) = 14.697, p < .001, partial η2 = .087. Follow-up analyses indicated that adolescents were more likely to view the physical and sexual violence conditions as dating violence, relative to the psychological and cyber-psychological conditions. Adolescents in the cyber-sexual condition were also more likely to perceive the situation as dating violence, relative to the cyber-psychological condition.
Perceptions of how seriously students felt that school administrators would consider the dating violence also differed across conditions, F(4, 629) = 28.30, p < .001, partial η2 = .15. As shown in Table 4, school administrators were viewed as significantly more likely to take both sexual and cyber-sexual dating violence more seriously than reports of physical, psychological, or cyber-psychological dating violence. Further, students rated school administrators as more likely to take physical dating violence more seriously than either psychological or cyber-psychological dating violence, with no significant differences between the latter two.
4.1.4 Perceptions of victims
Significant main effects of condition were indicated for ratings of how responsible the victim was for the dating violence, F(4, 625) = 17.63, p < .001, partial η2 = .10. The victim was rated as most responsible in the cyber-sexual scenario, compared to all other dating violence conditions. Victims in the sexual and cyber-psychological scenarios were also perceived as more responsible than the victim in the physical scenario. A similar main effect of condition was found to be significant for ratings of victim blame, F(4, 621) = 9.68, p < .001, partial η2 = .06. Victims in the cyber-sexual condition were seen as more blameworthy than those in the physical and psychological conditions, and victims of sexual and cyber-psychological violence were seen as more blameworthy than victims of physical violence.
Regarding perceptions of how upset the victim would be, there was a significant main effect of vignette condition, F(4, 627) = 10.15, p < .001, partial η2 = .06. Follow-up analyses showed that students rated the victim as more upset in the cyber-sexual condition, compared to the psychological, sexual, and cyber-psychological conditions, although the cyber-sexual and physical conditions did not differ significantly. Students in the physical and psychological conditions rated the victim as more upset than students in the cyber-psychological condition. Significant condition effects were also observed for ratings of how much the victim's future romantic relationships would be negatively affected as a result of the dating violence, F(4, 626) = 7.82, p < .001, partial η2 = .05. Students rated the dating violence as most negatively impacting the victim's future romantic relationships in the cyber-sexual condition, compared to the physical, psychological, and cyber-psychological conditions, with no significant differences between the cyber-sexual and sexual conditions. Students in the sexual condition rated the victim as being more negatively impacted compared to the psychological condition.
No significant differences were observed across types of dating violence for perceptions of how much the victim enjoyed the attention she received from the perpetrator, F(4, 621) = 2.04, p = .087, partial η2 = .0, with low ratings across conditions (i.e., less than 2.5 on a 7-point scale).
4.1.5 Perceptions of perpetrator
No significant differences were observed for perceptions of how responsible the perpetrator was for the dating violence, F(4, 626) = 1.95, p = .100, partial η2 = .01. Across conditions, students considered the perpetrator to be highly responsible for the dating violence, with all means above 6 on the 7-point scale. However, significant condition effects were observed regarding perceptions of whether the perpetrator should be punished for his actions, F(4, 624) = 31.18, p < .001, partial η2 = .17. Follow-up analyses showed that the perpetrator was rated as deserving more severe punishment in the physical, sexual, and cyber-sexual conditions, relative to the psychological and cyber-psychological conditions. There were no significant differences in ratings of punishment between the physical, sexual, and cyber-sexual conditions, nor between the psychological and cyber-psychological conditions.
Significant condition effects were observed for whether the perpetrator's violent behavior was because he really liked the victim, F(4, 614) = 4.27, p = .002, partial η2 = .03. Students in the psychological and cyber-psychological conditions were more likely to attribute the perpetrator's violent behavior to liking the victim than were students in the cyber-sexual condition.
Finally, significant variations were observed regarding whether adolescents thought that the perpetrator did not intend to harm the victim, but got carried away, F(4, 619) = 4.65, p = .001, partial η2 = .03. Students in the cyber-psychological condition were more likely to view the perpetrator as getting carried away, compared to the physical and cyber-sexual conditions.
4.2 Analyses for intentions to respond items and factors
A separate series of eight 2 (gender) by 2 (age) by 5 (vignette condition) ANOVAs were conducted to explore how adolescents thought they would respond to a peer disclosure of dating violence, as assessed by two subscales and six individual items. Responses were rated on 5-point, Likert scales, with higher scores reflecting greater likelihood of taking action. Given the number of analyses conducted, an adjusted alpha level of .006 was used to better control for type I error, with alphas between .006 and .011 reflecting marginally significant trends. Across analyses, all two-and three-way interactions were nonsignificant, and thus are not reported. Main effects for gender and age are presented first, followed by condition, for which follow-up pairwise analyses (Bonferroni) were conducted to detect significant differences across conditions.
4.2.1 Gender main effects
Results indicated significant gender main effects in four of the eight analyses conducted. Relevant means and standard deviations across gender and age groups are presented in Table 5. As shown, boys and girls did not differ significantly in how likely they would be to reach out to an adult for support or advice, talk to their own friends, talk to the perpetrator, or let the victim sort it out with the perpetrator. However, girls were significantly more likely to encourage or support the victim, encourage the victim to reach out to a friend, or to talk to their own parents. Boys were significantly more likely to indicate that they would do nothing.
| Subscale/Item (1–5 response scale) | Gender differences | Age differences | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| F(df) | p | Partial η2 | Boys | Girls | F(df) | p | Partial | Younger | Older | |
| (not at all likely–very likely) | M (SD) | M (SD) | η2 | M (SD) | M (SD) | |||||
| Subscales: | ||||||||||
| Reach out to adult for support | F(1, 624) = 0.856 | .355 | <.01 | 2.75 (1.04) | 2.83 (0.95) | F(1, 624) = 12.22 | <.001* | .02 | 2.93 (1.01) | 2.65 (.94) |
| Encourage or support Anna | F(1, 622) = 52.28 | <.001* | .08 | 4.22 (0.69) | 4.54 (0.42) | F(1, 622) = 0.37 | .545 | <.01 | 4.40 (0.53) | 4.37 (0.56) |
| Individual items: | ||||||||||
| Encourage Anna to reach out to friend(s) | F(1, 632) = 8.09 | .005* | .01 | 3.88 (1.16) | 4.13 (0.99) | F(1, 632) = 2.11 | .147 | <.01 | 3.94 (1.10) | 4.07 (1.00) |
| Talk to your own friend(s) | F(1, 628) = 3.261 | .071 | <.01 | 2.85 (1.21) | 3.04 (1.28) | F(1, 630) = 0.01 | .934 | <.01 | 2.94 (1.30) | 2.95 (1.21) |
| Talk to own parent/guardian | F(1, 630) = 33.33 | <.001* | .05 | 2.66 (1.24) | 3.28 (1.30) | F(1, 630) = 0.213 | .645 | <.01 | 3.00 (1.31) | 2.95 (1.31) |
| Do nothing | F(1, 626) = 18.37 | <.001* | .03 | 2.01 (1.04) | 1.65 (0.97) | F(1, 626) = 0.44 | .508 | <.01 | 1.86 (1.08) | 1.80 (0.93) |
| Speak to Liam | F(1, 632) = 1.74 | .187 | <.01 | 3.68 (1.19) | 3.55 (1.19) | F(1, 632) = 1.36 | .245 | <.01 | 3.67 (1.20) | 3.56 (1.18) |
| Let Anna sort it out with Liam | F(1, 629) = 2.51 | .114 | <.01 | 2.79 (1.07) | 2.65 (1.03) | F(1, 629) = 3.93 | .048 | .01 | 2.63 (0.98) | 2.81 (1.10) |
- Note: Significant mean differences at p < .006 are indicated with *.
4.2.2 Age main effects
Only one significant main effect of age was observed (see Table 5). Younger students (age 14–15) were more likely than older students (age 16–17) to reach out to an adult for support; older students were (marginally) more likely to talk to their own friends.
4.2.3 Differences across types of dating violence
Results indicated significant differences across dating violence conditions in only two of the eight analyses conducted, as described below. Means and standard deviations for each item are presented in Table 6.
| Subscale/Item (1–5 response scale) (not at all likely–very likely) | Type of dating violence | Significant post-hoc effects | ||||
|---|---|---|---|---|---|---|
| PHYS | PSYC | SEX | C-PSYC | C-SEX | ||
| M (SD) | M (SD) | M (SD) | M (SD) | M (SD) | ||
| (n = 136–139) | (n = 126–130) | (n = 117–119) | (n = 144–147) | (n = 116–118) | ||
| Subscale scores: | ||||||
| Reach out to adult for support | 2.86 (0.94) | 2.54 (0.92) | 2.99 (0.96) | 2.63 (1.03) | 2.93 (1.00) | Sex > Psyc, C-P C-S > Psyc |
| Encourage or support Anna | 4.42 (0.50) | 4.35 (0.54) | 4.44 (0.62) | 4.28 (0.54) | 4.43 (0.52) | All nonsignificant |
| Individual items: | ||||||
| Encourage Anna to reach out to friend(s) | 4.11 (0.95) | 4.01 (1.07) | 3.83 (1.18) | 4.13 (0.94) | 3.94 (1.13) | All nonsignificant |
| Talk to your own friend(s) | 3.28 (1.18) | 2.92 (1.22) | 2.69 (1.31) | 3.09 (1.29) | 2.73 (1.23) | Phys > Sex, C-S |
| Talk to own parent/guardian | 3.27 (1.27) | 2.78 (1.27) | 2.96 (1.34) | 3.11 (1.25) | 2.74 (1.40) | †Phys > Psyc, C-S |
| Do nothing | 1.77 (0.97) | 1.93 (1.06) | 1.75 (1.01) | 2.00 (1.11) | 1.69 (0.85) | All nonsignificant |
| Speak to Liam | 3.59 (1.23) | 3.41 (1.16) | 3.65 (1.23) | 3.62 (1.15) | 3.79 (1.20) | All nonsignificant |
| Let Anna sort it out with Liam | 2.71 (0.97) | 2.76 (1.02) | 2.63 (1.10) | 2.93 (1.02) | 2.57 (1.09) | All nonsignificant |
- Note: Significant differences are bolded. C-P, cyber-psychological; C-S, cyber-sexual; PHYS, physical; PSYC, psychological; SEX, sexual.
- Marginally significant trends are indicated with †.
A significant main effect of condition, F(4, 624) = 4.48, p = .001, partial η2 = .03, was obtained in the ANOVA exploring variations in the five-item composite assessing how likely students would be to reach out to an adult for support and advice after a dating violence disclosure. Follow-up analyses showed that students in the sexual condition were significantly more likely to seek adult support than students in either the psychological or cyber-psychological conditions, with nonsignificant differences between the latter two. Students in the cyber-sexual condition were more likely to reach out to an adult than students in the psychological condition.
Nonsignificant condition effects were observed for the six-item composite assessing how likely students would be to support or encourage the victim after the disclosure, F(4, 622) = 2.04, p = .088, partial η2 = .01. Across conditions, students rated the likelihood of supporting or encouraging the victim after the disclosure highly, with means above 4 on the five-point scale. Similarly, nonsignificant condition effects were observed for ratings of how likely students would be to encourage the victim to reach out to a friend after disclosing the incident, F(4, 632) = 1.67, p = .145, partial η2 = .01, with relatively high ratings (3.8 on the 5-point scale) across conditions.
A significant main effect of condition was found for how likely adolescents would be to talk to their own friends after a dating violence disclosure, F(4, 628) = 4.49, p = .001, partial η2 = .03. Students in physical violence condition were more likely to talk to their own friends than those in the sexual or cyber-sexual conditions. A similar, but marginally significant condition effect was observed for how likely participants would be to talk to their own parents after receiving the disclosure, F(4, 630) = 4.00, p = .009, partial η2 = .02, with greater likelihood of students talking to their own parents in the physical compared to psychological and cyber-sexual conditions.
Finally, nonsignificant condition effects were observed in the ANOVAs exploring variations in how likely adolescents would be to do nothing, F(4, 626) = 1.97, p = .097, partial η2 = .01, to speak to the perpetrator themselves, F(4, 632) = 1.41, p = .231, partial η2 = .01, or to let the victim sort it out with the perpetrator, F(4, 629) = 2.06, p = .085, partial η2 = .01.
5 DISCUSSION
The purpose of the present study was to expand the existing literature on adolescent responses to disclosures of dating violence. Specifically, we assessed whether high school students' perceptions of blame, understandings of violence, and intentions to respond to peer disclosures of dating violence would be influenced by participant age and gender, and the type of violence experienced. The effects observed were additive (main effects), not interactive, suggesting that age, gender, and the type of violence experienced each influenced adolescent attitudes toward and reactions to such peer disclosures. Further, this study was unique in assessing in-person forms of dating violence as well as the more recently recognized cyber forms.
Although further discussion of results are included below, here we present the overall main findings of the current study. Regarding age differences, younger students were more likely to view victims as blameworthy and to attribute the perpetrator's actions to liking the victim, and were also more likely to seek adult support following a disclosure of dating violence. Regarding participant gender, boys overall viewed all victims of dating violence as more responsible and blameworthy, and were also more likely to rate the perpetrator's behavior as more likely motivated by “liking the victim” or as just getting “carried away,” compared to girls. Further, in response to dating violence, boys were more likely to do nothing, compared to girls, however, were also less likely to provide encouragement or support to the victim. Considering type of dating violence, participants were more likely to consider physical and sexual aggression as incidents of dating violence than psychological forms (whether in-person or cyber forms of dating violence). Overall, the victim in the cyber-sexual scenario was considered as most responsible and blameworthy for her own victimization, relative to the other dating violence scenarios.
Across participant characteristics and type of violence portrayed in the current study, blame attributed to the victim and perpetrator varied. However, the overall perceptions of explicit blame toward the victim were relatively low and blame toward the perpetrator was consistently high. Even though participants rated victim blame and responsibility low, there was still some level of blame attributed to the victim. It is important to underscore the fact that any level of victim blaming has the potential to harm and hinder victim recovery. Victims who receive negative or blaming reactions report significant impacts to their well-being and recovery (Jackson et al., 2000; Moe, 2007). For victims who have not yet disclosed their assault, hearing others engage in victim blaming can send the message that victims are responsible for such violence, further discouraging reporting or seeking support for fear of being blamed. Ultimately, when considering victim blaming attitudes, it is important to note that even low levels of victim blaming have the potential to harm those who have experienced dating violence.
Considering our first hypothesis, and with regard to age, prior research has indicated that younger students are more likely to be accepting and tolerant of dating violence, compared to their older peers (e.g., Courtain & Glowacz, 2021; Price et al., 1999). In our study, considering a smaller age range, younger (14–15 years) and older students (16–17 years) were comparable in most of their reactions to disclosures dating violence. Younger students, however, were more likely to view victims as blameworthy and to attribute the perpetrator's actions to liking the victim, though they were also more likely to seek adult support following a disclosure of dating violence, consistent with previous qualitative work (Casey et al., 2018). With increasing age, comes greater experience with dating, romantic relationships, and/or dating violence which, in turn, may serve to decrease one's tolerance and acceptance of such behavior (e.g., Courtain & Glowacz, 2021; Price et al., 1999). The relative influences of age and dating experience in the formation of attitudes, beliefs, and reactions to dating violence remains a question for future research. Still, the age differences observed suggest that educational programs aimed at preventing dating violence must begin early.
Considering our second hypothesis and with regard to gender, although boys and girls did not differ in attributing blame to the perpetrator, boys viewed all victims of dating violence as more responsible, and their behavior as more blameworthy, than did girls. Boys were also more likely than girls to give the perpetrator, who was also a boy, the “benefit of the doubt,” rating victims as “liking the attention” from the perpetrator more, and rating the perpetrator's behavior as more likely motivated by “liking the victim” or as just getting “carried away.” Also, boys were less likely to encourage or support victims in response to dating violence disclosures, and less likely to encourage victims to talk to a friend, relative to girls. Boys were also less likely to talk to their parents, and more likely to do nothing in response to disclosure. Of note, the gender differences found within the current study are consistent with previous research with adult samples (e.g., Capezza & Arriaga, 2008; Dardis et al., 2017; Sylaska & Walters, 2014). Previous authors have explained gender differences in blame in terms of the defensive attribution hypothesis (Capezza & Arriaga, 2008; Grubb & Harrower, 2008), traditional gender roles (Dardis et al., 2017), and previous history with violence (Price et al., 1999), among other explanations.
A primary focus or our research was exploring differences in how adolescents perceive and respond to disclosures of different types of dating violence. Consistent with our third hypothesis, adolescents in the present sample were more likely to consider physical and sexual aggression as incidents of dating violence than psychological forms. Extending these findings to online dating violence, cyber-sexual violence was viewed as a form of dating violence more than cyber-psychological violence. Similarly, students felt that school administrators would take physical and sexual dating violence, both in-person and online, more seriously than psychological forms. These results likely reflect the perceived severity of the incident, regardless of whether in person or online. Similar results have also been reported for adult samples (e.g., Taylor & Sorenson, 2004).
Within the present study, participants may have perceived the victim's actions before the dating violence as adding legitimacy to that violence, especially with regard to online violence. For example, adolescents viewed the victim of cyber-sexual violence as most responsible for her victimization and more blameworthy, compared to other dating violence conditions. Students may have focused on information that the victim allowed the perpetrator to take a sexually explicit photo of her, rather than on information that the perpetrator shared that photo despite the victim's objections and expressions of discomfort. As another example, victims of sexual and cyber-psychological scenarios were seen as more responsible and blameworthy than victims of physical violence. Again, such perceptions may be attributable to victim actions before the assault. In this study, the victim in the sexual scenario initially started to kiss the perpetrator, although she soon withdrew her consent; the victim in the cyber-psychological scenario gave the perpetrator her phone and passwords. In contrast, the victim of physical violence was hit without provocation or consent. Although no victim should be blamed for the violence inflicted on them, past and present studies show that, for many participants, situational elements such as provocation or initial consent seem to blur the lines of blame and responsibility (e.g., Taylor & Sorenson, 2005), underscoring the need for prevention programs and bystander education to address such attention biases.
Regarding victim upset and impact, adolescents acknowledged that all victims would be upset by the dating violence experienced, and that the event would likely influence their future romantic relationships (i.e., means above 5/7 across conditions). Still, degree of upset and impact differed across types of dating violence. Not surprisingly, students viewed victims of physical violence to be more upset than victims of cyber-psychological violence, and victims of cyber-sexual violence to be more upset than victims of psychological violence, both in-person and online, likely reflecting the perceived severity of the events. Interestingly, cyber-sexual dating violence was viewed as more upsetting for the victim than psychological violence, both in-person and online, and was seen to have greater impact on future romantic relationships than physical or psychological violence (online and in-person). The findings suggest that adolescents do recognize that online aggression may be uniquely problematic for victims, with its potential for embarrassment across broad audiences and over time, its accompanying feelings of a loss of control over one's life, and its greater potential for damaging one's reputation relative to more isolated, in-person violence (Draucker & Martsolf, 2010; Stonard, 2020).
Adolescents in this study recognized the harm and upset that cyber-sexual violence would cause, yet they also viewed the victim of cyber-sexual violence as most responsible and most to blame for her victimization. This disconnect between perceptions of impact and consideration of blame may reflect a belief that, in certain contexts, the victim “should have known better” about a perpetrator's intentions (Wakelin & Long, 2003). Dating violence education programs need to address the mindset of victim blame, helping teens to recognize the many reasons why one may send a sexually explicit photo or stay in an abusive relationship and find ways to help or support, rather than blame the victim in these situations.
Perceptions of victim blame and responsibility must also be understood relative to perceptions of perpetrator blame and responsibility. In our study, adolescents viewed perpetrators as highly responsible for all types of dating violence (Means were 6/7 across conditions), and afforded victims much less responsibility (Means of 1.98 to 3.66 out of 7 across conditions). Although perpetrator responsibility was consistent across conditions, victim blame and responsibility were context-dependent, differing significantly across conditions.
Despite perpetrators being viewed as primarily responsible, some teens endorsed mitigating factors that serve to excuse or justify perpetrator behavior, especially for psychological forms of dating violence. Specifically, students were more likely to view psychological and cyber-psychological violence as a reflection of the perpetrator liking the victim, relative to cyber-sexual dating violence. They also viewed cyber-psychological violence as more likely to reflect that the perpetrator “just got carried away,” relative to perpetrators in the physical and cyber-sexual conditions. Qualitative research suggests that some forms of psychological dating violence, such as jealous or monitoring behaviors, are not necessarily considered abusive by youth, but are viewed as a normal part of relationship maintenance or demonstrate caring for one's partner (e.g., Draucker & Martsolf, 2010; Lucero et al., 2014). Although subtle, these forms of victim blaming can redistribute perceptions of blame and responsibility across victim and perpetrator, contributing to differential perceptions of the punishment warranted. In our study, perpetrators of psychological dating violence, both online and in-person, were seen as less deserving of punishment than perpetrators of physical, sexual, or cyber-sexual violence.
Overall, adolescent responses to dating violence disclosures were similar across conditions, with significant differences observed in only two of the eight responses considered. The most highly endorsed responses were efforts to encourage or support victims and encourage victims to reach out to their friends. Supportive and encouraging behavior has been found to be a common response to dating violence disclosures among teens in qualitative research (Casey et al., 2018), even among those who held victim-blaming attitudes. The least endorsed response, across conditions, was to “do nothing,” which was more often endorsed by boys than girls. Such a passive response, however, may not necessarily reflect a lack of caring, as Casey et al. (2018) found that many adolescents generally do want to intervene, but feel unequipped to do so safely and effectively in ways that actually support victims of dating violence. Future prevention efforts must help teens to identify positive and effective supporting behaviors.
Adolescent efforts to seek adult support in response to dating violence disclosures were more likely for sexual dating violence than psychological/cyber-psychological dating violence, and also more in response to cyber-sexual violence relative to psychological violence. This finding is consistent with evidence that adults are more likely to reach out to formal supports (e.g., police, counselors) when dating violence was considered more severe (e.g., sexual violence; Dardis et al., 2017; Taylor & Sorenson, 2004), although the present study considered a broader range of adults as formal sources of support than previous research. Youth do appear to acknowledge the serious nature of sexual and cyber-sexual dating violence, recognizing that outside adult support, such as school staff or the victim's parents, may be needed to adequately deal with such violence.
In the only other significant difference observed, students were more likely to reach out to their own friends following a disclosure of physical dating violence, as compared to disclosures of sexual or cyber-sexual violence. Considering the potentially visible harm engendered by physical victimization, perhaps teens in the present study felt greater need to seek advice in responding to such situations. Of interest in future research would be an exploration of what kinds of advice or support they seek from others, and what kinds of advice are actually helpful.
Finally, across conditions, a fair number of students endorsed the response of talking to the perpetrator (means of 3.41 to 3.79 out of 5 across conditions). In understanding this response, qualitative research has shown that some participants consider intervening or talking to the perpetrator to include expressing disapproval, or setting expectations for appropriate behaviors within a relationship; for others, it includes starting a physical altercation with the perpetrator (e.g., Casey et al., 2018). Future research should consider an even broader range of formal and informal response options to further our understandings of how students respond to such disclosures.
5.1 Limitations
Although results from the current study contribute to our understanding of perceptions and response to peer disclosures of teen dating violence, this study is not without limitations. First, participant perceptions and responses to dating violence were assessed using hypothetical scenarios which may not reflect what participants would actually do in that situation (see Felming et al., 2020; Waterman et al., 2021). Although research that asks about individual's lived experiences or provides open-ended responses is important, quantitative studies with hypothetical vignettes and forced-choice questions are also critical, allowing for direct comparisons of how certain factors influence one's reactions across different types of dating violence.
Another limitation of the current study concerns the gender of the victim and perpetrator. The authors acknowledge that dating violence can occur within all forms of relationships regardless of age, gender identity, or sexual orientation. It is important for future research within adolescent dating violence to explore these areas of study, as previous research has indicated that adolescent boys certainly experience dating violence victimization (sometimes at equal or higher rates than girls, depending on type) (Wincentak et al., 2016), and that noncisgender and nonheterosexual youth often experience higher (Espelage et al., 2018; Garthe et al., 2021) rates of dating violence victimization and perpetration than their cisgender and heterosexual counterparts. Finally, the current study explored responses to disclosures from “a peer she has come to know over the past year” across all conditions. We recognize, however, that the nature of the relationship one has with the victim can also impact responses and perceptions, as can one's own experience, or even one's knowledge of close friends or family members experiencing dating violence. Future research would benefit from consideration of these factors.
6 CONCLUSIONS
Despite these limitations, the present study fills an important gap in the literature by exploring how adolescents perceive and respond to peer dating violence, verifying findings from qualitative studies of teens, and extending research on victim-blaming in adults. In particular, this was the first study to explore differences in perceptions of in-person versus cyber forms of dating violence, and was unique in exploring adolescent perceptions. The present results underscore the need to consider how reactions and responses vary as a function of the type of violence that occurred, something that has been neglected in previous research, especially in regard to cyber forms of dating violence. Our results provide initial evidence that cyber forms of dating violence may engender greater victim blaming compared to in-person forms, particularly cyber-sexual dating violence in the context of victim responsibility, blame, impact, and upset. Hopefully, these findings can encourage further research on cyber aggression and its impact on victims, particularly in light of the greater reliance on digital technology to communicate and maintain relationships over the past 3 years of the COVID-19 pandemic.
Finally, the present findings serve to inform dating violence intervention and prevention programs, emphasizing the need to consider how different types of dating violence impact attitudes toward victims, perpetrators, understandings of violence, and intentions to respond. Research has shown that adolescent bystanders want to help victims of violence, but may lack the self-efficacy, knowledge, and skills to do so in an effective and safe way. Our results emphasize the need to address the specific perceptions and issues that are unique to each type of dating violence, rather than using a general approach that treats all forms of dating violence similarly. Importantly, both research and educational programs must give greater consideration to how cyber forms of dating violence may differ from in-person forms, better reflecting the realities of adolescents in the contemporary world.
ACKNOWLEDGMENTS
This research was supported by the Edith Lando Charitable Foundation, the Canadian network for Promoting Relationships and Eliminating Violence (PREVNet), the Public Health Agency of Canada, and participating schools. Katheryn Morrison was also supported in part by funding from the Social Sciences and Humanities Research Council. The authors wish to thank the adolescents who participated in this study.
CONFLICT OF INTEREST STATEMENT
The authors declare no conflict of interest.
ETHICS STATEMENT
This research received ethical approval from the University of British Columbia, Queen's University, McGill University, Grenfell Campus—Memorial University of Newfoundland, as well as all participating school districts. Informed consent and assent were obtained from parents/guardians and adolescent participants, respectively.
APPENDIX A
Dating Violence Vignette Examples:
The following are all the dating violence vignettes used in the present study. All vignettes included the same introduction, and then differed in their description of the dating violence that occurred.
Introduction for all Vignettes: Anna is a girl in your grade that you've gotten to know over the past year while working on several group projects. After school, Anna asks if you have time to talk with her. You agree and together you go to an empty classroom where you can talk privately. Anna tells you about something that happened earlier that day with Liam, a boy who she's been dating for 6 months. Anna tells you that she, Liam, and a few friends planned to skip class today and meet in the old band room to hang out, since the room is often empty. While the friends went to buy lunch down the street, Anna and Liam waited alone together in the band room.
Physical Scenario: Liam started asking Anna about how much time she's been spending with another boy from school. Even though Anna said that she is working on a school project with the boy and nothing more, Liam accused Anna of lying and cheating on him. Anna tried to tell Liam that she would never cheat on him, but Liam became angry and slapped Anna across the face, saying “stop lying to me.” Without saying anything more, Liam suggested they should leave and find their friends. As Anna tells you the story, she seems confused about what happened, and what she should do.
Psychological Scenario: Liam started asking Anna about where she had been yesterday, as Liam had called and texted her over 10 times last night and Anna never picked up or responded to Liam's calls and texts. When Anna told Liam that she had gone to bed early because she did not feel well, Liam started yelling at Anna, calling her a piece of trash and accusing her of lying and cheating. Anna denied cheating, but Liam just started yelling more, saying that Anna was worthless and lucky to be with him. Without saying anything more, Liam suggested they should leave and find their friends. As Anna tells you the story, she seems confused about what happened, and what she should do.
Sexual Scenario: Liam started to kiss Anna, and at first Anna kissed Liam back. Liam told Anna how much he cared about her, and Anna said the same to him. Then Liam started to fondle and undress Anna. Anna felt nervous about the situation and asked Liam to stop, saying she was nervous and worried their friends might walk in and see them. Liam kept telling Anna how much he liked her and that their friends wouldn't be back for a while. Liam told Anna that if she really cared, she would want to show him how much. Anna said she didn't want to keep going, but Liam proceeded to have sex with her. When Liam finished, they got dressed and left to find their friends. As Anna tells you the story, she seems confused about what happened, and what she should do.
Cyber-Sexual Scenario: Liam started asking Anna to take off her clothes so Liam could take a “sexy” photo of her. Anna felt nervous and said no, but Liam kept telling Anna how much he cared for her, and that if Anna cared she would do this for him. Anna still felt unsure but was tired of saying no, so she agreed to take the photo if Liam promised that it was only for him and not to be shared. Liam agreed and took the photo, but later in the day Anna saw Liam sharing the photo with his friends. As Anna tells you the story, she seems confused about what happened, and what she should do.
Cyber-Psychological Scenario: Liam started complaining that Anna seemed more interested in her social media than in spending time with him. Anna denied spending a lot of time on social media, but that she liked to keep up with her friends online. Liam then took Anna's phone and went through her message history. Liam asked for Anna's passwords so he could check on Anna's social media from his phone, saying “if you trust me, you'll give me your passwords.” Anna reluctantly gave Liam her passwords. Liam then set up an app on Anna's phone so he could track her location anytime, which Liam said was to protect Anna. Without saying anything more, Liam suggested they should leave and find their friends. As Anna tells you the story, she seems confused about what happened, and what she should do.
Open Research
DATA AVAILABILITY STATEMENT
Data are available upon reasonable request from the authors.
REFERENCES
- 1 Although the authors do not view gender as a boy/girl binary, previous research on attitudes toward dating violence and victim blaming have treated gender as such, excluding those who do not fit within it.
- 2 Analyses regarding the impact of sexual orientation among gay/lesbian students were conducted for all dependent variables. Specifically, 17 gay or lesbian students were matched by gender, age group, school location, and vignette condition with 17 heterosexual students. Results of independent t-tests, conducted with an adjusted alpha of .005, were all nonsignificant.
