Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Dec 2.
Published in final edited form as: J Interpers Violence. 2022 Nov 11;38(9-10):6576–6600. doi: 10.1177/08862605221135166

Middle School Psychological Distress and Sexual Harassment Victimization as Predictors of Dating Violence Involvement

Gabriel R Murchison 1,2, S Bryn Austin 3,4,5, Sari L Reisner 6,7,8,9, Jarvis T Chen 3
PMCID: PMC10692829  NIHMSID: NIHMS1942697  PMID: 36367156

Abstract

Adolescent sexual harassment victimization is increasingly recognized as a strong risk factor for dating violence victimization and perpetration. Research on this association has focused on older adolescents and on sexual harassment at a single time point rather than chronic exposure. Furthermore, potential mechanisms, such as psychological distress, are not well understood. The goals of this study were to identify whether sexual harassment victimization and psychological distress were reciprocally related, whether chronic psychological distress and sexual harassment victimization in early adolescence were associated with higher levels of dating violence involvement measured in mid-adolescence, and whether these relationships differed between boys and girls. We used longitudinal data from 4,718 U.S. middle school students to fit regression models for the associations between sexual harassment victimization and psychological distress. We then used data from a subset of 1,279 students followed up in high school to fit marginal structural models for 6th-8th grade psychological distress and sexual harassment victimization as predictors of dating violence involvement measured in 9th grade. We found that (1) sexual harassment victimization was positively, concurrently associated with psychological distress, with a stronger association among girls than boys; (2) psychological distress was positively, prospectively associated with sexual harassment victimization among girls but not boys; (3) chronic 6th-8th grade psychological distress was not significantly associated with dating violence victimization or perpetration measured in 9th grade; and (4) chronic 6th-8th grade sexual harassment victimization was associated with significantly higher levels of dating violence victimization and perpetration measured in 9th grade. The results support a reciprocal relationship between distress and sexual harassment victimization for early adolescent girls, and they call for further attention to gender differences in the content and impact of sexual harassment. Furthermore, the findings indicate that preventing sexual harassment in early adolescence may be critical in reducing dating violence.

Keywords: Sexual harassment, dating violence, adolescence, gender, psychological distress

Dating violence (i.e., physical, sexual, and/or psychological aggression by a dating partner) is a considerable threat to adolescent well-being, associated with elevated long-term risk of depressive symptomatology, suicidality, and substance use (Banyard & Cross, 2008; Exner-Cortens et al., 2013). Among the strongest predictors of dating violence victimization and perpetration are other forms of victimization, particularly peer victimization (Boivin et al., 2011; Hamby et al., 2012; Hébert et al., 2019). One relevant form of peer victimization is peer sexual harassment: unwelcome verbal, visual, or physical sexual behaviors (Hill & Kearl, 2011) directed at an adolescent by similar-age peers. Adolescents who experience more peer sexual harassment victimization report more dating violence involvement as targets (Chiodo et al., 2009; Gagné et al., 2005; Hamby et al., 2012; Hequembourg et al., 2020) and perpetrators (Boivin et al., 2011). Indeed, compared to other forms of peer victimization, sexual harassment victimization is a stronger predictor of dating violence involvement (Hamby et al., 2012; Hébert et al., 2019; Hequembourg et al., 2020). Improved understanding of how peer sexual harassment victimization contributes to dating violence is needed to identify adolescents in need of targeted interventions—particularly those at risk for dating violence perpetration—and illuminate risk processes that universal prevention programs should address.

Sexual Harassment and Dating Violence from a Polyvictimization Perspective

One way of conceptualizing the association between sexual harassment victimization and dating violence victimization is in terms of polyvictimization: the co-occurrence of multiple forms of victimization against the same individual, often across multiple perpetrators and settings (Finkelhor, 2008). Polyvictimization research tends to focus on identifying general risk factors linked to multiple forms of victimization. One notable risk factor of this type is psychological distress, i.e., emotional suffering marked by depression and anxiety symptoms, which may be accompanied by somatic symptoms such as headaches (Drapeau et al., 2012). Psychological distress may both reflect and exacerbate social isolation, leaving adolescents with less peer or adult support to protect them from victimization (Finkelhor, 2008). Psychological distress may also impair perception and avoidance of risky situations, potentially causing aggressors to perceive distressed adolescents as easier targets (Curry & Youngblade, 2006; Finkelhor, 2008). Furthermore, distressed adolescents may be more likely to provoke or escalate conflicts, potentially because distress and its antecedent stressors have interfered with the development of adaptive emotion regulation and conflict resolution strategies (Doyle & Sullivan, 2017; Segrin, 2000). In sum, the polyvictimization literature suggests that psychological distress may put adolescents at risk for dating violence victimization, and for distressed adolescents who provoke or escalate conflicts, there may be an elevated risk of dating violence perpetration.

More than a shared risk factor for these outcomes, psychological distress could represent a mediator between sexual harassment and dating violence. Sexual harassment victimization is a longitudinal predictor of psychological distress (Chiodo et al., 2009; Rinehart et al., 2017), and distressed adolescents are at elevated risk for dating violence perpetration and victimization (Foshee et al., 2001; Lehrer et al., 2006; Ulloa et al., 2014). Moreover, one study has found that greater distress mediates the association between sexual harassment victimization and dating violence perpetration (Boivin et al., 2011). Thus, it is plausible that sexual harassment victimization contributes to dating violence victimization and perpetration in part by increasing psychological distress. This proposed pathway is more notable, but also more complex, in light of studies suggesting a positive longitudinal association between psychological distress and sexual harassment victimization as well as vice versa (Dahlqvist et al., 2016; Marshall et al., 2013). These findings suggest that distress and sexual harassment victimization promote one another, a cycle that could further increase the risk of dating violence involvement. However, these studies had limited adjustment for potential confounders that could plausibly affect both psychological distress and sexual harassment victimization, including peer victimization, child abuse, and other family factors. Thus, it is not clear to what extent adolescents may experience a vicious cycle of psychological distress and sexual harassment victimization.

Sexual Harassment and Dating Violence from a Feminist Perspective

In addition to general risk factors such as psychological distress, sexual harassment and dating violence share specific risk factors related to gender and sexuality: patriarchal attitudes (Fineran & Bennett, 1999; Reyes et al., 2016; Rizzo et al., 2020), acceptance of sexual aggression (Collibee et al., 2019), and rigid, binary gender expectations (Conroy, 2013; Reyes et al., 2016; Rizzo et al., 2020). Drawing on these findings, feminist analyses of adolescent sexual harassment have postulated that this behavior enforces hegemonic masculinity (Conroy, 2013; Fineran & Bennett, 1999; Robinson, 2005), i.e., the dominance of men enacting socially privileged versions of masculinity over women, less gender-conforming men, and nonbinary people (Connell & Messerschmidt, 2005). Sexual harassment is thought to accomplish this in two distinct ways: (1) establishing boys’ power over girls and (2) denigrating those who violate hegemonic gender norms (Conroy, 2013). This analysis aligns with apparent gender differences in the character of adolescent sexual harassment. Sexual harassment targeting girls often involves unwanted sexual advances by boys, which girls perceive as an assertion of boys’ power over them (Hand & Sanchez, 2000; Hill & Kearl, 2011; Keddie, 2009; Robinson, 2005). Sexual harassment targeting boys often involves disparaging their gender expression, i.e., enforcement of gender norms (Chiodo et al., 2009; Conroy, 2013).

To the extent that sexual harassment reinforces male dominance and hegemonic gender norms, these experiences may be influential as adolescents begin to form dating relationships. For adolescent girls, male-dominance-enforcing sexual harassment could increase tolerance for boys’ aggressive behavior and willingness to date boys who exhibit aggression, although this issue has received relatively little research attention. For adolescent boys, gender-norm-enforcing sexual harassment victimization may result in preoccupation with being perceived as masculine; these boys may use aggression as a means of performing a more hegemonic masculinity (Rizzo et al., 2020). Even non-heterosexual relationships may be shaped by these heterosexual “scripts” (Kubicek et al., 2014). Girls’ dating violence perpetration and boys’ victimization fit less neatly into the framework discussed above. However, adolescent dating violence is often bidirectional, and aggressive behavior by one partner can lead to aggression by the other, either in self-defense or escalation (Foshee et al., 2007; Gray & Foshee, 1997). Thus, girls’ sexual harassment victimization could indirectly influence their dating violence perpetration by contributing to their dating violence victimization, while boys’ sexual harassment victimization could indirectly influence their dating violence victimization by contributing to their dating violence perpetration.

Gaps in the Literature

The polyvictimization and feminist perspectives bring to light gaps in the literature on adolescent sexual harassment and dating violence. One is a lack of adequate control for confounding by other forms of violence. While some prior studies of the relationship between sexual harassment victimization and dating violence involvement have adjusted for family violence exposure (Gagné et al., 2005) or other forms of peer victimization (Hequembourg et al., 2020), none to our knowledge have adjusted for both. Similarly, prior studies of the proposed bidirectional relationship between psychological distress and sexual harassment victimization have had very limited confounder adjustment (Dahlqvist et al., 2016; Marshall et al., 2013).

A second gap is that gender differences in the processes linking sexual harassment victimization, psychological distress, and dating violence involvement have not been well characterized. When adolescent girls and boys are asked to report how sexual harassment affected them (e.g., how upset they felt when it happened), girls typically characterize these experiences as more distressing and disruptive (Fineran & Bennett, 1999; Hill & Kearl, 2011; Mitchell et al., 2014). However, when sexual harassment victimization has been modeled as a predictor of general psychological distress, some studies have found a stronger association among girls than boys (Dahlqvist et al., 2016) while others have found comparable associations in boys and girls (Chiodo et al., 2009; Rinehart et al., 2017). One potential explanation for this pattern is that adolescent boys are less comfortable acknowledging the impact of sexual harassment due to gender norms that discourage boys from displaying vulnerability (Chu, 2005), yet the differences in general psychological distress associated with sexual harassment victimization may be similar for boys and girls. Alternately, sexual harassment may indeed cause more general distress for girls than boys, perhaps due to differences in the types of sexual harassment that girls and boys tend to experience (Crowley & Cornell, 2020; Espelage et al., 2016; Hand & Sanchez, 2000; Hill & Kearl, 2011) or because girls experience more pervasive sexual objectification and intimidation than boys (Calogero et al., 2020). Further evidence is needed to clarify whether there are gender differences in the overall (rather than directly-attributed) psychological impact of sexual harassment victimization. In a similar vein, feminist perspectives (discussed above) suggest that sexual harassment victimization could have a particularly strong effect on boys’ dating violence perpetration and girls’ dating violence victimization. However, few studies have addressed gender differences in these associations, with mixed results for victimization (Chiodo et al., 2009; Hamby et al., 2012) and little evidence of gender-based effect modification for perpetration (Boivin et al., 2011). Additional studies are needed to clarify whether sexual harassment victimization is in fact differentially associated with dating violence victimization or perpetration for girls compared to boys.

A third gap is a lack of studies addressing the roles of chronic psychological distress and sexual harassment victimization in relation to dating violence. Within the peer victimization literature, chronic peer victimization—that is, frequent peer victimization in each of several measurement periods (e.g., school years)—has been linked to more negative psychosocial outcomes compared to frequent peer victimization during some measurement periods but not others (Bowes et al., 2013; Kochenderfer-Ladd & Wardrop, 2001). However, prior studies of sexual harassment victimization in relation to dating violence involvement have typically considered victimization during a single school year (Chiodo et al., 2009; Hequembourg et al., 2020) or lifetime victimization (any/none; Hamby et al., 2012) rather than estimating the impact of chronic victimization. Similarly, it is plausible that psychological distress would need to be sustained over time to substantively interfere with protective social relationships or impede development of emotion regulation and conflict resolution skills, yet prior studies of the association between psychological distress and dating violence involvement have not estimated the potential impact of chronically elevated distress across multiple school years.

A final gap is limited prior attention to sexual harassment in early adolescence. Early adolescence is marked by the emergence of mixed-gender peer groups and early dating relationships (Connolly et al., 2004), suggesting that sexual harassment during this period may have a marked impact on expectations for male-female interactions and sexual/romantic relationships. Nonetheless, previous research addressing associations between sexual harassment victimization and dating violence involvement has primarily been conducted with older adolescents, leaving the impact of early adolescent sexual harassment victimization unclear.

The Present Study

The present study was intended to address the gaps discussed above. Our first goal was to determine whether the apparent reciprocal relationship between early adolescent psychological distress and sexual harassment victimization is robust to adjustment for key potential confounders. Our second goal was to assess the association between chronic early adolescent psychological distress and dating violence involvement, and our third goal was to assess the association between chronic early adolescent sexual harassment victimization and dating violence involvement. To these ends, we tested the following hypotheses:

  • H1. Higher levels of sexual harassment victimization during a given year of middle school will be associated with higher levels of psychological distress in the same year.

  • H2. Higher levels of psychological distress during a given year of middle school will be associated with higher levels of sexual harassment victimization in the following year.

  • H3. Higher levels of psychological distress in 6th, 7th, and 8th grade will be associated with higher levels of dating violence involvement in 9th grade.

  • H4. Higher levels of sexual harassment victimization in 6th, 7th, and 8th grade will be associated with higher levels of dating violence involvement in 9th grade.

An additional goal was to test whether these associations varied for boys compared to girls. In keeping with previous research, discussed above, we hypothesized that the association between sexual harassment victimization and concurrent psychological distress would be stronger among girls than boys. Following from a feminist analysis of sexual harassment, also discussed above, we hypothesized that sexual harassment victimization would be more strongly associated with dating violence victimization among girls, while sexual harassment victimization would be more strongly associated with dating violence perpetration among boys.

Methods

Data

We conducted a secondary analysis of longitudinal data from a cluster-randomized trial of a social-emotional learning intervention, Second Step (Espelage & Bub, 2018; Espelage et al., 2015). Baseline data were collected in the fall of 2010, when participants were in 6th grade (mean age=11 years). Follow-up data were collected annually in the spring terms of 2011 (6th grade) through 2016 (11th grade), for a total of 7 waves over 6 school years. The first 5 waves (4 school years) are analyzed here; we used data from later waves for certain demographic variables. All measures were based on student questionnaires completed during the school day.

The full sample comprised 4,718 students who attended 36 participating middle schools in Kansas and Illinois. Both Kansas and Illinois students were included in analyses for H1 and H2, which required only 6th-8th grade data. Illinois students who continued into participating high schools (n=1,452) were also surveyed in 9th through 11th grades. Because dating violence involvement was not assessed until 9th grade, analyses for H3 and H4 were restricted to Illinois students. Students who had never dated (n=173; 11.9%) were excluded from H3 and H4 analyses, resulting in a final subsample size of n=1,279 for the H3 and H4 analyses. Table 1 presents descriptive statistics for the full sample and the H3/H4 subsample.

Table 1.

Characteristics of 4,718 participating students in Kansas (H1 and H2 analyses only) and Illinois (all analyses)

Full sample (all participating Kansas and Illinois students), n=4,718 Subsample for H2 and H3 (Illinois participants who had dated by Spring 9th grade), n=1,279
n (%) n (%)
Gender
  Female 2167 (47.9) 511 (49.9)
  Male 2359 (52.1) 513 (50.1)
Sexual orientation identity
  Heterosexual – 899 (87.9)
  Sexual minority – 124 (12.1)
Race/ethnicity
  White 426 (24.6) 189 (24.5)
  Black or African American 648 (37.3) 273 (35.4)
  Hispanic or Latino/a/x 429 (24.7) 215 (27.9)
  Asian or Asian American 13 (0.7) 6 (0.8)
  Biracial or multiracial 219 (12.6) 88 (11.4)
Ever dated by spring 7th grade
  No 799 (26.4) 146 (18.8)
  Yes 2222 (73.6) 631 (81.2)
Ever dated by spring 8th grade
  No 605 (21.4) 86 (10.2)
  Yes 2217 (78.6) 755 (89.8)
Ever dated by spring 9th grade
  No – 0 (0.0)
  Yes – 1024 (100.0)
History of physical abuse at baseline
  No 2630 (81.0) 600 (84.2)
  Yes 617 (19.0) 113 (15.8)
History of sexual abuse at baseline
  No 2738 (84.7) 682 (95.9)
  Yes 495 (15.3) 29 (4.1)
Had witnessed domestic abuse at baseline
  No 3086 (95.4) 612 (86.2)
  Yes 149 (4.6) 98 (13.8)
Second Step implementation
  Attended school that implemented Second Step 2038 (45.0) 458 (44.7)
  Attended school that did not implement Second Step 2488 (55.0) 566 (55.3)
Open in a new tab

Note. Variables assessed only during the high school waves are presented for the dating violence analysis subsample only. Counts do not sum to sample sizes due to wave and item missingness.

Students gave voluntary assent and were informed that responses were confidential unless they disclosed intent to harm themselves or being harmed by an adult. Parents/guardians received letters about the study and instructions for opting students out. The original study was approved by participating school districts and the investigators’ institutional review board. The Harvard T. H. Chan School of Public Health Institutional Review Board determined that our analysis was exempt from review.

Measures

Psychological distress and sexual harassment victimization.

Psychological distress and sexual harassment victimization were measured at baseline and 6th-, 7th, and 8th-grade follow-up. We operationalized chronic psychological distress and chronic sexual harassment victimization as above-mean scores on these variables at all three follow-up waves. Psychological distress was measured with 8 items based on the Modified Depression Scale (Dunn et al., 2012; sample: “In the last 30 days, how often… Were you very sad?”; 1=Never; 5=Almost always). Cronbach’s alpha was .86–.90 across waves. Sexual harassment victimization was measured using an index of 9 behaviors (sample: “Touched, grabbed, or pinched you in a sexual way”; American Association of University Women Educational Foundation & Harris Interactive, 2001). Students reported how often “OTHER KIDS [have] done the following things to YOU when you did not want them to” (1=Never; 4=10 or more times). Instructions referred to lifetime events at baseline and events “since the start of school” at follow-up. Cronbach’s alpha was .81–.88 across waves.

Dating violence victimization and perpetration.

Dating violence victimization and perpetration were measured in 9th grade using the victimization and perpetration scales (25 items each) of the Conflict in Adolescent Dating Relationships Inventory (CADRI; Wolfe et al., 2001), assessing the frequency of sexual abuse, verbal/emotional abuse, relational abuse, threats, and physical abuse (1=Never; 4=Often [6 or more times]). Instructions refer to past and present relationships with “a boyfriend/girlfriend.” Cronbach’s alpha=0.91 (victimization) and 0.88 (perpetration). The CADRI has demonstrated acceptable test-retest reliability and interrater (dating partner) agreement in a high school sample (Wolfe et al., 2001).

Demographic variables and potential confounders.

Student gender was coded as male or female based on the most frequent self-reported gender across waves. (No student consistently reported a nonbinary gender across waves.) Race/ethnicity was based on students’ self-report at baseline, selected from the options “American Indian or Alaska Native,” “African American/Black,” “Asian,” “Hispanic/Latino,” “White,” and “Pacific Islander.” Sexual orientation identity was assessed at waves 5 through 7. The latest non-missing response was used, and responses other than “heterosexual or straight” were recoded to “sexual minority.”

Parental monitoring was measured at baseline using a modified version of the Poor Family Management scale from Arthur et al. (2002); Cronbach’s alpha=0.86. Family functioning was assessed at baseline using the 12-item General Functioning subscale of the McMaster Family Assessment Device (Epstein et al., 1983); Cronbach’s alpha=0.80. Child abuse exposure was assessed at baseline using three items assessing physical abuse victimization, sexual abuse victimization, and witnessing domestic abuse. During initial modeling, we used indicator variables for each form of abuse and found similar coefficients for each. Consequently, for parsimony, these forms of abuse were summed to form an index of child abuse victimization.

Bullying victimization was measured at baseline and 6th-, 7th, and 8th-grade follow-up with three items from the University of Illinois Bully Scale (Espelage & Holt, 2001). Students reported how often they experienced bullying events in the last 30 days (sample: “Other students called me names”; 0=Never; 4=7 or more times). Cronbach’s alpha was 0.89–0.92 across waves.

As discussed above, this study was a secondary analysis of data from a cluster-randomized trial of Second Step, which consisted of 41 classroom lessons intended to reduce student aggression and substance use, including perpetration of bullying, cyberbullying, homophobic teasing, and sexual harassment (Espelage et al., 2015). Models were adjusted for an indicator of whether a student’s middle school implemented Second Step.

Missing Data

There was substantial attrition (37.3%) from baseline to 8th grade, which differed significantly by gender, race/ethnicity, baseline sexual harassment victimization, Second Step implementation, family functioning, parental monitoring, and witnessing domestic abuse. Among the high school subsample, 16.3% did not participate in the 9th grade survey. Item missingness was greatest in early waves and highest for the three child abuse items (12.2-12.6%). By wave 5, item missingness was under 2% for most items and highest (up to 4.6%) for CADRI items. Abuse and CADRI items were near the ends of the surveys, and missingness was comparable to the items directly preceding them, suggesting that missingness was not primarily due to the sensitive nature of the questions. Missingness was addressed by multiple imputation (m=50) using the mice package for R (Van Buuren & Groothuis-Oudshoorn, 2011). Differential attrition was addressed by imputing the data in “wide” format (Asendorpf et al., 2014).

Analysis

Analyses were conducted using R 3.4.4 (R Core Team, 2017).

Models for reciprocal relationships between sexual harassment victimization and psychological distress (H1 and H2).

Psychological distress was measured with reference to a 30-day period whereas sexual harassment victimization was measured with reference to the entire school year. Consequently, we assumed that sexual harassment victimization causally preceded psychological distress within a given year. To determine whether sexual harassment victimization during middle school was concurrently associated with psychological distress, we fit a linear regression model for psychological distress as a function of sexual harassment victimization at the same time point (dichotomized at the mean for consistency with the marginal structural models, discussed below), adjusted for sexual harassment and psychological distress at the previous time points (Z scores), bullying victimization at the same and previous time points (Z scores), gender (binary), race/ethnicity (categorical), baseline family functioning (Z score), baseline parental monitoring (Z score), child abuse victimization (index), and Second Step implementation (binary). The model also included an interaction term between gender and psychological distress. Time points were treated as separate observations; generalized estimating equations with an autoregressive covariance structure were used to account for within-participant correlation. This model used exposure, time-varying confounder, and outcome data from the spring of 6th, 7th, and 8th grades. To determine whether psychological distress during middle school was prospectively associated with sexual harassment victimization, we fit a linear regression model like the one described above, modeling sexual harassment victimization as a function of psychological distress (dichotomized at the mean) at the previous time point. This model used exposure and time-varying confounder data from the spring of 6th and 7th grades and outcome data from the spring of 7th and 8th grades.

Models for dating violence involvement as a function of chronic psychological distress and sexual harassment victimization throughout middle school (H3 and H4).

Traditional multivariable regression, as used for the short-term exposure models described above, is often unsuitable for estimating the joint effect of exposure at multiple time points (here, chronic exposure over 3 years). This is because variables may act as mediators for one exposure time point but confounders for subsequent time points (“time-varying confounding”). Regression adjustment for time-varying confounders induces mediator adjustment bias for the first exposure, while non-adjustment confounds estimates for subsequent exposures. Marginal structural models (MSMs) overcome this dilemma by addressing time-varying confounders with weighting rather than regression adjustment (VanderWeele et al., 2016). We used MSMs to test the dating violence hypotheses because several covariates (e.g., bullying victimization) were likely time-varying confounders. Methodological details for these models are provided in Appendix A. Because MSM estimates for continuous exposures are generally unstable, psychological distress and sexual harassment victimization were dichotomized at the mean (VanderWeele et al., 2016).

Results

Table 1 presents counts and proportions for all categorical variables. Means and standard deviations for continuous variables are presented in Appendix B, Supplemental Table B1.

Table 2 presents estimated differences in psychological distress for students with above-mean (versus below-mean) sexual harassment victimization during the same year (H1). Among girls, above-mean sexual harassment victimization was associated with elevated psychological distress scores during the same year, with a mean difference of b=2.12 (95% CI: 1.74, 2.50; P<0.001). This association was significantly weaker for boys (interaction term p=0.032), but still significant, with a mean difference of b=1.57 (95% CI: 1.22, 1.92; p<0.001).

Table 2.

Gender-specific estimates and linear regression coefficients for the association between 6th/7th/8th-grade sexual harassment victimization and concurrent psychological distress, as predicted by current-year sexual harassment victimization and potential confounders, among 4,718 middle school students in Kansas and Illinois

Gender-specific estimates b (95% CI) p
Association between above-mean sexual harassment victimization and concurrent psychological distress score, girls 2.12 (1.74, 2.50) <0.001
Association between above-mean sexual harassment victimization and concurrent psychological distress score, boys 1.57 (1.22, 1.92) <0.001
Linear regression coefficients for concurrent psychological distress b (95% CI) p
Sexual harassment victimization above mean, time t 2.12 (1.74, 2.50) <0.001
Male x sexual harassment victimization above mean, time t −0.55 (−1.04, −0.06) 0.032
Sexual harassment victimization, time t-1 (z score) −0.18 (−0.30, −0.05) 0.008
Time-varying potential confounders
Psychological distress, time t-1 (z score) 3.68 (3.51, 3.85) <0.001
Bullying victimization, time t (z score) 2.00 (1.83, 2.18) <0.001
Bullying victimization, time t-1 (z score) −0.76 (−0.93, −0.60) <0.001
Gender (ref=female)
  Male −2.03 (−2.27, −1.79) <0.001
Race/ethnicity (ref=White)
  Black or African American −0.30 (−0.59, −0.02) 0.046
  Hispanic or Latino/a/x 0.26 (0.04, 0.49) 0.023
  Asian or Asian American 0.24 (−0.71, 1.19) 0.636
  Biracial or multiracial 0.11 (−0.36, 0.58) 0.653
Baseline family context variables
Family functioning - moderate (ref=low) −0.12 (−0.25, 0.00) 0.066
Parental monitoring (z score) −0.04 (−0.16, 0.08) 0.501
Child abuse victimization index 0.04 (−0.14, 0.22) 0.681
Second Step implementation (ref=school did not implement Second Step)
  Attended school that implemented Second Step 0.06 (−0.15, 0.26) 0.586
Open in a new tab

Note. Estimates are based on a repeated-measures linear generalized estimating equations model. Gender-specific estimates were calculated using the main effect and interaction terms for sexual harassment victimization from the linear regression model.

Table 3 presents estimated differences in sexual harassment victimization for students with above-mean (versus below-mean) psychological distress during the previous year (H2). For girls, above-mean psychological distress was associated with greater sexual harassment victimization during the following year, b=0.25 (95% CI: 0.03, 0.47; p=0.030; Table 3). For boys, above-mean psychological distress was not significantly associated with sexual harassment victimization during the following year, b=0.11 (95% CI: −0.16, 0.39; p=0.440).

Table 3.

Gender-specific estimates and linear regression coefficients for the association between 6th/7th-grade psychological distress and subsequent-year sexual harassment victimization among 4,718 middle school students in Kansas and Illinois

Gender-specific estimates b (95% CI) p
Association between above-mean psychological distress and subsequent-year sexual harassment victimization score, girls 0.25 (0.03, 0.47) 0.030
Association between above-mean psychological distress and subsequent-year sexual harassment victimization score, boys 0.11 (−0.16, 0.39) 0.440
Linear regression coefficients for subsequent-year sexual harassment victimization b (95% CI) p
Psychological distress above mean, time t 0.25 (0.03, 0.47) 0.030
Male x psychological distress above mean, time t −0.14 (−0.48, 0.20) 0.435
Psychological distress, time t-1 (z score) 0.03 (−0.10, 0.16) 0.633
Time-varying potential confounders
Sexual harassment victimization, time t (z score) 1.24 (1.01, 1.46) <0.001
Sexual harassment victimization, time t-1 (z score) 0.44 (0.25, 0.64) <0.001
Bullying victimization, time t (z score) 0.02 (−0.14, 0.19) 0.797
Bullying victimization, time t-1 (z score) 0.04 (−0.07, 0.16) 0.477
Gender (ref=female)
  Male −0.28 (−0.45, −0.12) 0.001
Race/ethnicity (ref=White)
  Black or African American 0.23 (−0.04, 0.49) 0.116
  Hispanic or Latino/a/x −0.27 (−0.45, −0.09) 0.006
  Asian or Asian American 0.14 (−0.36, 0.65) 0.580
  Biracial or multiracial 0.22 (−0.11, 0.54) 0.212
Baseline family context variables
Family functioning (z score) −0.03 (−0.12, 0.06) 0.489
Parental monitoring (z score) 0.04 (−0.03, 0.12) 0.280
Child abuse victimization index 0.00 (−0.13, 0.13) 0.969
Second Step implementation (ref=school did not implement Second Step)
  Attended school that implemented Second Step program −0.11 (−0.25, 0.03) 0.145
Open in a new tab

Note. Estimates are based on a repeated-measures linear generalized estimating equations model. Gender-specific estimates were calculated using the main effect and interaction terms for psychological distress from the linear regression model.

Table 4 presents estimated differences in dating violence involvement as a function of psychological distress (H3) and sexual harassment victimization (H4), using marginal structural models to estimate the effects of above-mean distress/victimization in 6th, 7th, and 8th grades (i.e., chronic exposure) relative to below-mean distress/victimization in 6th, 7th, and 8th grades. Chronic above-mean psychological distress was not significantly associated with dating violence victimization, b=2.31 (95% CI: 0.16, 4.45; p=0.057), or perpetration, b=0.79 (95% CI: −0.33, 1.90; p=0.174). However, chronic above-mean sexual harassment victimization was associated with higher levels of dating violence victimization, b=4.34 (95% CI: 2.41, 6.27; p<0.001), and perpetration, b=1.82 (95% CI: 0.58, 3.06; p=0.006). Other coefficients, including effect estimates for each year, are presented in Appendix B. In all models, effect estimates were largest for 8th grade and smallest for 6th grade. There was no evidence of effect modification by gender.

Table 4.

Joint estimates for associations between 6th, 7th, and 8th grade exposures (psychological distress; sexual harassment victimization) and dating violence involvement through 9th grade among 1,279 middle/high school students in Illinois

b (95% CI) p
Mean difference in dating violence victimization associated with above-mean psychological distress (ref = below mean), 6th, 7th, and 8th grades 2.31 (0.16, 4.45) 0.057
Mean difference in dating violence perpetration associated with above-mean psychological distress (ref = below mean), 6th, 7th, and 8th grades 0.79 (−0.33, 1.90) 0.174
Mean difference in dating violence victimization associated with above-mean sexual harassment victimization (ref = below mean), 6th, 7th, and 8th grades 4.34 (2.41, 6.27) <0.001
Mean difference in dating violence perpetration associated with above-mean sexual harassment victimization (ref = below mean), 6th, 7th, and 8th grades 1.82 (0.58, 3.06) 0.006
Open in a new tab

Note. Estimates are based on marginal structural models. Stabilized inverse probability of treatment weights were used to adjust for time-varying confounding by sexual harassment victimization and bullying victimization. Regression was used to adjust for time-invariant confounding by 6th grade bullying and sexual harassment victimization, gender, sexual orientation identity, race/ethnicity, baseline family functioning, child abuse victimization, and whether the adolescent’s school implemented the Second Step social-emotional learning program.

Discussion

Early adolescence is a key period for social-emotional development regarding gender, sexuality, and romantic relationships. The present study extends prior research on the associations among sexual harassment victimization, psychological distress, and dating violence by considering chronic exposure throughout early adolescence while adjusting for key potential confounders, notably child maltreatment, family factors, and bullying victimization. The results contribute further evidence that sexual harassment victimization is an important risk factor for dating violence victimization and build on limited evidence that it is also a risk factor for dating violence perpetration (Boivin et al., 2011). The results also add nuance to previous evidence of a reciprocal relationship between psychological distress and sexual harassment victimization, showing that the distress-victimization association was present for girls but not boys and that the victimization-distress association was stronger for girls than boys. Contrary to hypotheses, there was no significant association between chronic early adolescent psychological distress and dating violence involvement, nor did gender modify the association between chronic early adolescent sexual harassment victimization and dating violence involvement.

While prior studies of the association between sexual harassment victimization and dating violence have generally measured sexual harassment during a single year of high school, the present study extends this association to chronic sexual harassment victimization in early adolescence. This finding reinforces the importance of preventing peer sexual harassment during middle school. Multicomponent school- and community-based interventions have emerged as a preferred approach to adolescent violence prevention, and at least two such programs—Shifting Boundaries (Taylor et al., 2013) and Dating Matters® (DeGue et al., 2021)—have been shown to reduce both sexual harassment and dating violence in middle school students. Shifting Boundaries incorporates classroom lessons about sexual harassment and dating violence, a poster campaign, temporary school-based restraining orders, and an activity in which students map harassment hot spots where adult supervision is needed (Taylor et al., 2013). Dating Matters® focuses on healthy relationships skills and comprises classroom lessons, parent education, school staff training, a youth “brand ambassador” component, and a community-level assessment and planning component (DeGue et al., 2021). Given the key role of gender norms in sexual harassment, it is important to note that gender is not a central focus of either Shifting Boundaries or Dating Matters®, though both address gender roles to some extent. It may be advantageous to combine these programs with curricula that address gender norms more comprehensively.

We did not identify significant associations between chronic early adolescent psychological distress and dating violence perpetration or victimization. This finding was surprising given previous research identifying psychological distress as a risk factor for dating violence involvement (Foshee et al., 2001; Lehrer et al., 2006; Ulloa et al., 2014). For dating violence victimization, the estimate was near-significant, and it is possible that our analysis was under-powered. Alternatively, the prior findings could be explained by confounders addressed in the present study (e.g., bullying victimization). A third possibility is that the previously observed effects are short-term ones, while the present study focused on long-term effects. This scenario would be consistent with a previous finding that depressed affect and dating violence perpetration were positively associated at the same time point but not at 18-month follow-up (Foshee et al., 2001). Micro-longitudinal studies could be valuable in exploring this possibility.

Turning to the hypothesized reciprocal relationship between psychological distress and sexual harassment victimization, we found support for both directions only among girls. Among boys, there was no evidence that psychological distress was associated with greater risk of sexual harassment victimization, and the concurrent association between sexual harassment victimization and psychological distress was significantly weaker for boys than girls. These gender differences may be related to differences in the types of sexual harassment that girls and boys tend to experience; for instance, girls disproportionately experience forced sexual contact (e.g., pinching, kissing) and sexual rumor-spreading, whereas boys disproportionately experience homophobic name-calling and having clothes pulled or removed (Crowley & Cornell, 2020; Espelage et al., 2016; Hand & Sanchez, 2000; Hill & Kearl, 2011). Because schools and peer groups may penalize forced sexual contact more so than verbal or visual sexual harassment, perpetrators of forced sexual contact (typically against girls) may selectively target peers they perceive as vulnerable, such as those higher in psychological distress, while perpetrators of verbal and visual harassment may be less selective. Furthermore, both girls and boys rate forced sexual contact and sexual rumor-spreading as particularly distressing forms of sexual harassment (Crowley & Cornell, 2020; Hand & Sanchez, 2000; Hill & Kearl, 2011), so their disproportionate prevalence among girls may contribute to the stronger association between sexual harassment victimization and distress for girls compared to boys.

Despite some gender differences, we found further evidence linking sexual harassment victimization to internalizing symptoms in both boys and girls (Espelage et al., 2016; Hill & Kearl, 2011). Moreover, our results suggest that both boys and girls may respond to sexual harassment victimization with externalizing behaviors such as dating violence perpetration. Thus, school staff should be trained to recognize and address sexual harassment against students of all genders, including behaviors disproportionately targeting one gender. Given the strong association between psychological distress and sexual harassment victimization, middle school mental health personnel (e.g., counselors) should consider screening students presenting with psychological distress for sexual harassment victimization. With students’ permission, mental health personnel can then coordinate with other staff to intervene against future harassment.

Limitations & Methodological Strengths

This study has several notable limitations. First, because this was a secondary analysis and data on dating violence prior to 9th grade was not available, we could not adjust for baseline dating violence involvement. Similarly, we could not rule out feedback between dating violence and 6th-8th grade sexual harassment victimization or psychological distress, a form of reverse causation that could confound causal interpretations of the direct effects of psychological distress and sexual harassment victimization on dating violence involvement. Second, some sexual harassment could have been a form of dating violence, though prior studies suggest this overlap is negligible (Hamby et al., 2012; Hequembourg et al., 2020). Third, as in most observational studies, unmeasured confounding could have biased the results. Fourth, we addressed gender differences only between boys and girls due to a dearth of nonbinary participants, yet nonbinary students may experience frequent and severe sexual harassment (Mitchell et al., 2014). Finally, sexual harassment reflects not only gendered power dynamics but also intersecting systems of power relations, notably racism (Harris & Kruger, 2020); we could not address this crucial intersection because the present dataset lacked measures of interpersonal or structural racism.

This study also had notable methodological strengths. We used a detailed measure of psychological, physical, and sexual dating violence involvement. We analyzed five waves of data spanning four developmentally important years for sexual harassment and dating violence, and we adjusted for key potential confounders, including family factors and bullying victimization. Moreover, we used marginal structural models to address time-varying confounding, a technique that should be considered for future research addressing victimization at multiple time points. Furthermore, we addressed each research question in both boys and girls, identifying similarities and differences between these groups.

Conclusion

Sexual harassment is increasingly recognized as a key risk factor for adolescent dating violence. The present study found that students who experience chronic sexual harassment in early adolescence are at greater risk of dating violence perpetration and victimization. Results highlight a potential problematic cycle of psychological distress and sexual harassment victimization; gender differences in these findings call for further attention to differences in boys’ and girls’ sexual harassment experiences. Our findings make clear that early adolescence is a key period for sexual harassment and dating violence prevention. We suggest that middle schools implement multicomponent sexual harassment/dating violence programs (DeGue et al., 2021; Taylor et al., 2013) and implement curricula dispelling harmful gender norms. We also suggest that school mental health personnel screen distressed students for sexual harassment victimization. These strategies have the potential to protect adolescents’ mental health while reducing their risk of dating violence involvement.

Supplementary Material

Appendix

Acknowledgments

The authors wish to thank Dr. Dorothy Espelage and Dr. Kristen Bub for making their Second Step trial data available for secondary use. They also thank the staff of the Inter-university Consortium for Political and Social Research for facilitating access to the dataset.

Funding

SB Austin is supported by US Maternal and Child Health Bureau, Health Resources and Services Administration grant T76-MC00001. GR Murchison is supported by National Institute of Mental Health grant T32-MH020031.

Footnotes

Conflicts of Interest

None to declare.

References

  1. American Association of University Women Educational Foundation, & Harris Interactive. (2001). Hostile hallways: Bullying, teasing, and sexual harassment in school. https://files.eric.ed.gov/fulltext/ED454132.pdf
  2. Arthur MW, Hawkins JD, Pollard JA, Catalano RF, & Baglioni AJ (2002). Measuring risk and protective factors for substance use, delinquency, and other adolescent problem behaviors: The Communities That Care Youth Survey. Evaluation Review, 26(6), 575–601. 10.1177/019384102237850 [DOI] [PubMed] [Google Scholar]
  3. Asendorpf JB, Van De Schoot R, Denissen JJA, & Hutteman R (2014). Reducing bias due to systematic attrition in longitudinal studies: The benefits of multiple imputation. International Journal of Behavioral Development, 38(5), 453–460. 10.1177/0165025414542713 [DOI] [Google Scholar]
  4. Banyard VL, & Cross C (2008). Consequences of teen dating violence: Understanding intervening variables in ecological context. Violence Against Women, 14(9), 998–1013. 10.1177/1077801208322058 [DOI] [PubMed] [Google Scholar]
  5. Boivin S, Lavoie F, Hébert M, & Gagné M-H (2011). Past victimizations and dating violence perpetration in adolescence: The mediating role of emotional distress and hostility. Journal of Interpersonal Violence, 27(4), 662–684. 10.1177/0886260511423245 [DOI] [PubMed] [Google Scholar]
  6. Bowes L, Maughan B, Ball H, Shakoor S, Ouellet-Morin I, Caspi A, Moffitt TE, & Arseneault L (2013). Chronic bullying victimization across school transitions: The role of genetic and environmental influences. Development and Psychopathology, 25(2), 333–346. 10.1017/S0954579412001095 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Calogero RM, Tylka TL, Siegel JA, Pina A, & Roberts T-A (2020). Smile pretty and watch your back: Personal safety anxiety and vigilance in objectification theory. Journal of Personality and Social Psychology. 10.1037/pspi0000344 [DOI] [PubMed] [Google Scholar]
  8. Chiodo D, Wolfe DA, Crooks C, Hughes R, & Jaffe P (2009). Impact of sexual harassment vctimization by peers on subsequent adolescent victimization and adjustment: A longitudinal study. Journal of Adolescent Health, 45(3), 246–252. 10.1016/j.jadohealth.2009.01.006 [DOI] [PubMed] [Google Scholar]
  9. Chu JY (2005). Adolescent boys’ friendships and peer group culture. New Directions for Child and Adolescent Development, 2005(107), 7–22. 10.1002/cd.118 [DOI] [PubMed] [Google Scholar]
  10. Collibee C, Rizzo C, Bleiweiss K, & Orchowski LM (2019). The influence of peer support for violence and peer acceptance of rape myths on multiple forms of interpersonal violence among youth. Journal of Interpersonal Violence. 10.1177/0886260519832925 [DOI] [PubMed] [Google Scholar]
  11. Connell RW, & Messerschmidt JW (2005). Hegemonic masculinity: Rethinking the concept. Gender & Society, 19(6), 829–859. 10.1177/0891243205278639 [DOI] [Google Scholar]
  12. Connolly J, Craig W, Goldberg A, & Pepler D (2004). Mixed-gender groups, dating, and romantic relationships in early adolescence. Journal of Research on Adolescence, 14(2), 185–207. 10.1111/j.1532-7795.2004.01402003.x [DOI] [Google Scholar]
  13. Conroy NE (2013). Rethinking adolescent peer sexual harassment: Contributions of feminist theory. Journal of School Violence, 12(4), 340–356. 10.1080/15388220.2013.813391 [DOI] [Google Scholar]
  14. Crowley BZ, & Cornell D (2020). Associations of bullying and sexual harassment with student well-being indicators. Psychology of Violence, 10(6), 615–625. 10.1037/vio0000345 [DOI] [Google Scholar]
  15. Curry LA, & Youngblade LM (2006). Negative affect, risk perception, and adolescent risk behavior. Journal of Applied Developmental Psychology, 27(5), 468–485. 10.1016/j.appdev.2006.06.001 [DOI] [Google Scholar]
  16. Dahlqvist HZ, Landstedt E, Young R, & Gådin KG (2016). Dimensions of peer sexual harassment victimization and depressive symptoms in adolescence: A longitudinal cross-lagged study in a Swedish sample. Journal of Youth and Adolescence, 45(5), 858–873. 10.1007/s10964-016-0446-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. DeGue S, Niolon PH, Estefan LF, Tracy AJ, Le VD, Vivolo-Kantor AM, Little TD, Latzman NE, Tharp A, Lang KM, & Taylor B (2021). Effects of Dating Matters® on sexual violence and sexual harassment outcomes among middle school youth: a cluster-randomized controlled trial. Prevention Science, 22(2), 175–185. 10.1007/s11121-020-01152-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Doyle ST, & Sullivan TN (2017). Longitudinal relations between peer victimization, emotion dysregulation, and internalizing symptoms among early adolescents. Journal of Early Adolescence, 37(2), 165–191. 10.1177/0272431615594458 [DOI] [Google Scholar]
  19. Drapeau A, Marchand A, & Beaulieu-Prévost D (2012). Epidemiology of psychological distress. In L’Abate L (Ed.), Mental illnesses – Understanding, prediction and control (pp. 105–134). InTech. [Google Scholar]
  20. Dunn EC, Johnson RM, & Green JG (2012). The Modified Depression Scale (MDS): A brief, no-cost assessment tool to estimate the level of depressive symptoms in students and schools. School Mental Health, 4(1), 34–45. 10.1007/s12310-011-9066-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Epstein NB, Baldwin LM, & Bishop DS (1983). The McMaster family assessment device. Journal of Marital and Family Therapy, 9(2), 171–180. 10.1111/j.1752-0606.1983.tb01497.x [DOI] [Google Scholar]
  22. Espelage DL, & Bub KL (2018). Effects of a middle school social-emotional learning program on bullying, teen dating violence, sexual violence, and substance use in high school, Illinois, 2010-2016 [Data set]. Inter-university Consortium for Political and Social Research. 10.3886/ICPSR36726.v1 [DOI] [Google Scholar]
  23. Espelage DL, & Holt MK (2001). Bullying and victimization during early adolescence. Journal of Emotional Abuse, 2(2-3), 123–142. 10.1300/j135v02n02_08 [DOI] [Google Scholar]
  24. Espelage DL, Hong JS, Rinehart S, & Doshi N (2016). Understanding types, locations, & perpetrators of peer-to-peer sexual harassment in U.S. middle schools: A focus on sex, racial, and grade differences. Children and Youth Services Review, 71, 174–183. 10.1016/j.childyouth.2016.11.010 [DOI] [Google Scholar]
  25. Espelage DL, Low S, Van Ryzin MJ, & Polanin JR (2015). Clinical trial of Second Step middle school program: Impact on bullying, cyberbullying, homophobic teasing, and sexual harassment perpetration. School Psychology Review, 44(4), 464–479. 10.17105/spr-15-0052.1 [DOI] [Google Scholar]
  26. Exner-Cortens D, Eckenrode J, & Rothman E (2013). Longitudinal associations between teen dating violence victimization and adverse health outcomes. Pediatrics, 131(1), 71–78. 10.1542/peds.2012-1029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Fineran S, & Bennett L (1999). Gender and power issues of peer sexual harassment among teenagers. Journal of Interpersonal Violence, 14(6), 626–641. 10.1177/088626099014006004 [DOI] [Google Scholar]
  28. Finkelhor D. (2008). Childhood victimization: Violence, crime and abuse in the lives of young people. Oxford University Press. [Google Scholar]
  29. Foshee VA, Bauman KE, Linder F, Rice J, & Wilcher R (2007). Typologies of adolescent dating violence: Identifying typologies of adolescent dating violence perpetration. Journal of Interpersonal Violence, 22(5), 498–519. 10.1177/0886260506298829 [DOI] [PubMed] [Google Scholar]
  30. Foshee VA, Linder F, MacDougall JE, & Bangdiwala S (2001). Gender differences in the longitudinal predictors of adolescent dating violence. Preventive Medicine, 32(2), 128–141. 10.1006/pmed.2000.0793 [DOI] [PubMed] [Google Scholar]
  31. Gagné M-H, Lavoie F, & Hébert M (2005). Victimization during childhood and revictimization in dating relationships in adolescent girls. Child Abuse & Neglect, 29(10), 1155–1172. 10.1016/j.chiabu.2004.11.009 [DOI] [PubMed] [Google Scholar]
  32. Gray HM, & Foshee V (1997). Adolescent dating violence: Differences between one-sided and mutually violent profiles. Journal of Interpersonal Violence, 12(1), 126–141. 10.1177/088626097012001008 [DOI] [Google Scholar]
  33. Hamby S, Finkelhor D, & Turner H (2012). Teen dating violence: Co-occurrence with other victimizations in the National Survey of Children’s Exposure to Violence (NatSCEV). Psychology of Violence, 2(2), 111–124. 10.1037/a0027191 [DOI] [Google Scholar]
  34. Hand JZ, & Sanchez L (2000). Badgering or bantering?: Gender differences in experience of, and reactions to, sexual harassment among U.S. high school students. Gender & Society, 14(6), 718–746. 10.1177/089124300014006002 [DOI] [Google Scholar]
  35. Harris J, & Kruger AC (2020). “We always tell them, but they don’t do anything about it!” Middle school Black girls’ experiences with sexual harassment at an urban middle school. Urban Education. 10.1177/0042085920959131 [DOI] [Google Scholar]
  36. Hébert M, Daspe M-È, Lapierre A, Godbout N, Blais M, Fernet M, & Lavoie F (2019). A meta-analysis of risk and protective factors for dating violence victimization: The role of family and peer interpersonal context. Trauma, Violence, & Abuse, 20(4), 574–590. 10.1177/1524838017725336 [DOI] [PubMed] [Google Scholar]
  37. Hequembourg AL, Livingston JA, & Wang W (2020). Prospective associations among relationship abuse, sexual harassment and bullying in a community sample of sexual minority and exclusively heterosexual youth. Journal of Adolescence, 83, 52–61. 10.1016/j.adolescence.2020.06.010 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Hill C, & Kearl H (2011). Crossing the line: Sexual harassment at school. American Association of University Women. https://files.eric.ed.gov/fulltext/ED525785.pdf [Google Scholar]
  39. Keddie A. (2009). ‘Some of those girls can be real drama queens’: Issues of gender, sexual harassment and schooling. Sex Education, 9(1), 1–16. 10.1080/14681810802639863 [DOI] [Google Scholar]
  40. Kochenderfer-Ladd B, & Wardrop JL (2001). Chronicity and instability of children’s peer victimization experiences as predictors of loneliness and social satisfaction trajectories. Child Development, 72(1), 134–151. 10.1111/1467-8624.00270 [DOI] [PubMed] [Google Scholar]
  41. Kubicek K, McNeeley M, & Collins S (2014). “Same-sex relationship in a straight world”: Individual and societal influences on power and control in young men’s relationships. Journal of Interpersonal Violence, 30(1), 83–109. 10.1177/0886260514532527 [DOI] [PubMed] [Google Scholar]
  42. Lehrer JA, Buka S, Gortmaker S, & Shrier LA (2006). Depressive symptomatology as a predictor of exposure to intimate partner violence among US female adolescents and young adults. Archives of Pediatrics & Adolescent Medicine, 160(3), 270–276. 10.1001/archpedi.160.3.270 [DOI] [PubMed] [Google Scholar]
  43. Marshall SK, Faaborg-Andersen P, Tilton-Weaver LC, & Stattin H (2013). Peer sexual harassment and deliberate self-injury: Longitudinal cross-lag investigations in Canada and Sweden. Journal of Adolescent Health, 53(6), 717–722. 10.1016/j.jadohealth.2013.06.009 [DOI] [PubMed] [Google Scholar]
  44. Mitchell KJ, Ybarra ML, & Korchmaros JD (2014). Sexual harassment among adolescents of different sexual orientations and gender identities. Child Abuse & Neglect, 38(2), 280–295. 10.1016/j.chiabu.2013.09.008 [DOI] [PubMed] [Google Scholar]
  45. R Core Team. (2017). R: A language and environment for statistical computing (Version 3.4.4) [Computer software]. R Foundation for Statistical Computing. https://www.R-project.org [Google Scholar]
  46. Reyes HLM, Foshee VA, Niolon PH, Reidy DE, & Hall JE (2016). Gender role attitudes and male adolescent dating violence perpetration: Normative beliefs as moderators. Journal of Youth and Adolescence, 45(2), 350–360. 10.1007/s10964-015-0278-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Rinehart SJ, Espelage DL, & Bub KL (2017). Longitudinal effects of gendered harassment perpetration and victimization on mental health outcomes in adolescence. Journal of Interpersonal Violence, 35(23-24), 5997–6016. 10.1177/0886260517723746 [DOI] [PubMed] [Google Scholar]
  48. Rizzo AJ, Banyard VL, & Edwards KM (2020). Unpacking adolescent masculinity: Relations between boys’ sexual harassment victimization, perpetration, and gender role beliefs. Journal of Family Violence. 10.1007/s10896-020-00187-9 [DOI] [Google Scholar]
  49. Robinson KH (2005). Reinforcing hegemonic masculinities through sexual harassment: Issues of identity, power and popularity in secondary schools. Gender and Education, 17(1), 19–37. 10.1080/0954025042000301285 [DOI] [Google Scholar]
  50. Segrin C. (2000). Social skills deficits associated with depression. Clinical Psychology Review, 20(3), 379–403. 10.1016/S0272-7358(98)00104-4 [DOI] [PubMed] [Google Scholar]
  51. Taylor BG, Stein ND, Mumford EA, & Woods D (2013). Shifting Boundaries: An experimental evaluation of a dating violence prevention program in middle schools. Prevention Science, 14(1), 64–76. 10.1007/s11121-012-0293-2 [DOI] [PubMed] [Google Scholar]
  52. Ulloa EC, Martinez-Arango N, & Hokoda A (2014). Attachment anxiety, depressive symptoms, and adolescent dating violence perpetration: A longitudinal mediation analysis. Journal of Aggression, Maltreatment & Trauma, 23(6), 652–669. 10.1080/10926771.2014.920452 [DOI] [Google Scholar]
  53. Van Buuren S, & Groothuis-Oudshoorn K (2011). mice: Multivariate imputation by chained equations in R. Journal of Statistical Software, 45(3). 10.18637/jss.v045.i03 [DOI] [Google Scholar]
  54. VanderWeele TJ, Jackson JW, & Li S (2016). Causal inference and longitudinal data: A case study of religion and mental health. Social Psychiatry and Psychiatric Epidemiology, 51(11), 1457–1466. 10.1007/s00127-016-1281-9 [DOI] [PubMed] [Google Scholar]
  55. Wolfe DA, Scott K, Reitzel-Jaffe D, Wekerle C, Grasley C, & Straatman A-L (2001). Development and validation of the Conflict in Adolescent Dating Relationships Inventory. Psychological Assessment, 13(2), 277–293. 10.1037/1040-3590.13.2.277 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix
NIHMS1942697-supplement-Appendix.docx (5.4MB, docx)

RESOURCES