Adolescent and emerging adult women are at heightened risk of intimate partner violence (IPV) victimization, with 70% of women reporting their first experience with partner rape, physical abuse, or stalking occurred by the age of 24 (Black et al., 2011). Though this early relationship violence may be severe (Kennedy et al., 2018), young women may be reluctant to tell anyone what is happening (Kennedy et al., under review; Bundock et al., 2018) in part because their disclosure may be met with mixed reactions, including blame, disbelief, or otherwise unhelpful responses (Barter et al., 2009; Dworkin et al., 2019; Sullivan et al., 2010; Toscano, 2007; Wood et al., 2010). Nevertheless, disclosure is survivors’ first step toward receiving emotional support and material aid, and can play a key role in leaving the relationship, thus interrupting the abuse (Wood et al., 2010). Research on adolescents’ IPV disclosure is an emergent area. Key limitations of the adolescent disclosure literature include an overall lack of studies; the use of homogenous, school-based samples; an emphasis on description rather than an exploration of the factors shaping disclosure; and few qualitative studies, which can offer insight into IPV disclosure as a multi-determined, complex process over time (Kennedy et al., under review; Bundock et al., 2018). With such a meager empirical base, we are limited in our ability to develop evidence-based approaches for an increasingly diverse generation of adolescents. If we can build knowledge about what factors seem to facilitate or impede disclosure, we can more effectively address and prevent adolescent IPV—after all, if young women do not disclose or seek help, violence in these early relationships will remain hidden from view and beyond our interventions.

To build this knowledge, we need a strong theoretical understanding. Life course theory emphasizes the importance and timing of role transitions, stability and change in social trajectories over time, and the tenet that one’s life experiences are linked to others, with the family as the foundational context (Elder, 1998). Early experiences with victimization and adversity are understood to play a role in shaping later poor health, mental health, and economic outcomes (Hatch, 2005). From this perspective, an adolescent girl’s first relationship is critically important, as it represents a key transition from the family of origin to the broader trajectory of partners; abusive first relationships in particular merit careful scrutiny. By examining these first experiences, and the factors shaping disclosure, we have a window into the genesis and unfolding of relationship violence and the process by which young survivors attempt to make sense of their experiences and decide whether and how to seek support. Thus, in the current study we use grounded theory to explore IPV victimization (physical, coercive control, and sexual) during young women’s abusive relationships with their first boyfriends, and the factors that shape their process of disclosure, within a socioeconomically diverse sample recruited from a university, a two-year college, and community sites serving low-income young women.

The conceptual model developed by Kennedy et al. (2012a) complements life course theory and captures the complexities of women’s process of disclosing and attaining help for victimization. Though designed with formal support in mind, the model is well-suited to guide our exploration of young women’s IPV disclosure process to both informal supports (e.g., friends, family members) and formal providers, including law enforcement (LE) and health and mental health staff. The model foregrounds the importance of social location; prior victimization and exposure to violence and other adversity, both familial and extra-familial; and the community, developmental, and situational contexts in shaping disclosure and help attainment. Stigma and stigmatization also play a role. Seeking and attaining effective help begins with an appraisal of one’s needs, defining or labeling the problem, and assessing the availability of support and aid: If a young survivor does not define her experience as abusive, or the perceived disadvantages of disclosure outweigh the benefits, or she evaluates that she has no accessible sources of support, she is unlikely to disclose the IPV and seek help (Kennedy et al., 2012a).

Social location, which refers to one’s position within intersecting systems of stratification such as gender, socioeconomic status (SES), and race/ethnicity, fundamentally shapes the life course, influencing opportunities, exposure to cumulative adversity, and access to resources (Hatch, 2005; Pearlin, 1989). Poor women of color occupy a disadvantaged social location and as a result face increased risk of IPV—especially severe and lethal IPV—and are more likely to experience discrimination and stigma when they attempt to attain effective help, compared to their higher-SES White peers (Crenshaw, 1991; Goodman et al., 2009; Kennedy & Prock, 2018; West, 2004). Childhood victimization and exposure to violence are associated with IPV during adolescence, with cumulative, more severe victimization linked to more severe IPV victimization (Hamby et al., 2012; Kennedy et al., 2012b; Vézina & Hébert, 2007). How might prior victimization shape a young woman’s later IPV disclosure process? Witnessing her mother’s severe physical abuse may cause her to minimize her own abuse, or fail to define it as abusive, because she perceives her experience to be relatively less severe (Few & Rosen, 2005). Additionally, if disclosure of prior victimization, such as child sexual abuse, was met with parental disbelief or blame, she may be less likely to disclose later IPV (Smith & Cook, 2008).

Community factors also play a critical role in influencing IPV disclosure among women, particularly to formal providers. Poor, high-crime neighborhoods tend to have limited resources, including social service systems, coupled with sometimes-unresponsive, sometimes-harmful LE; both of these factors can negatively influence IPV survivors’ decision to disclose and attempt to obtain effective help (Kennedy et al., 2012a; Miller, 2008). The Kennedy et al. (2012a) model draws on life course theory in its understanding of the developmental factors that shape attaining help for IPV, emphasizing transition periods as times of heightened stress and vulnerability. For example, a girl in early adolescence who enters into a serious relationship may be at higher risk of abuse, compared to an older adolescent, as she is more likely to idealize the relationship, and less knowledgeable about what constitutes abuse; peer pressure may also play a key role (Barter et al., 2009; Sullivan et al., 2010; Wood et al., 2010). As a result, she may not label her experience as abusive, or may minimize it, and thus fail to disclose and attain effective help (Kennedy et al., 2018; Toscano, 2007; Wood et al., 2010).

Among survivors in general, situational factors that positively influence disclosure and seeking help include the type of IPV (physical vs. sexual), its severity, level of fear, presence of children, and, for sexual violence, the degree to which it conforms to dominant perceptions of “real” rape (i.e., a singular event involving force committed by a stranger) and thus is labeled as a rape (Kennedy et al., under review; Flicker et al., 2011; Kennedy et al., 2012a; Littleton et al., 2008; Sabina & Ho, 2014; Sylaska & Edwards, 2014). Coercive control may also play a key role, via the isolation tactics employed by abusive partners (Johnson, 2006; Øverlien et al., 2019; Stark, 2007). For adolescent girls especially, additional situational factors include their relationships with both family members and peers. Can a family member or friend be trusted to be supportive, caring, and helpful? Do peer group norms reflect and maintain compulsory heterosexuality (e.g., young men as biologically-driven sexual predators, Tolman et al., 2003)? Do the benefits of telling a friend outweigh the disadvantages, such as the loss of status or privacy that could result? Along with the general situational factors cited above, these appear especially salient in shaping adolescent women’s disclosure (Barter et al., 2009; Black et al., 2008; Smith & Cook, 2008; Sullivan et al., 2010; Tolman et al., 2003; Toscano, 2007; Wood et al., 2010).

Finally, stigma and stigmatization fundamentally influence disclosure. Stigma—which is imposed on survivors by others via victim blaming and negative, judgmental reactions to disclosure—facilitates stigmatization, understood as the internalization of stigma by survivors, in the form of self-blame, shame, and anticipatory stigma, i.e., the expectation that others will blame and judge a survivor for her abuse (Kennedy & Prock, 2018). Together, they are powerful barriers to disclosure, shaping how women define their experience, how they evaluate the benefits and drawbacks of disclosing and to whom, and how they are responded to, once they disclose. Survivors may blame themselves and thus resist labeling their experience as abusive, they may feel too much shame and embarrassment to tell anyone, or they may suspect that others will blame them, discount their abuse, or make the situation worse (Kennedy & Prock, 2018). For adolescent girls who are grappling with identity development while also seeking peer acceptance, self-blame, shame, and the fear of social embarrassment can be critical impediments to disclosing IPV (Amar & Alexy, 2005; Barter et al., 2009; Toscano, 2007).

For the present study, we sought to extend current findings, create new knowledge, and build theory on early IPV and disclosure among adolescent women, with a focus on their abusive first relationships. Drawing on the Kennedy et al. (2012a) model and using a grounded theory approach (Strauss & Corbin, 1990), we were guided by the following research questions: 1) How does IPV, including physical, coercive control, and sexual IPV, manifest during these abusive first relationships? 2) What does the process of disclosure/non-disclosure look like? 3) How do aspects of social location; prior victimization; community, developmental, and situational factors; and stigma and stigmatization interact in different ways to influence this process?

Method

Research Design, Recruitment, and Participants

The current study derives from a larger mixed-methods project, a cross-sectional interview design that examined IPV (physical, coercive control, sexual) at the relationship level within a diverse sample of 148 young women from a university (n = 50), a two-year college (n = 48), and community sites serving low-income young women (n = 50; Kennedy et al., 2018). Prior to recruitment, we received approval from our institutional review board. At each site, we distributed flyers describing a study on “partner conflict” using specific examples (e.g., being hit) in order to reach participants who may not have labeled their experiences as abusive. Eligibility criteria included young women between the ages of 18 and 24, who had experienced at least one of the three types of IPV by a male partner, and could read and speak English.

In the larger study, we used the life history calendar (LHC), a method demonstrated to improve accurate recall of violence over the life course in comparison to retrospective surveys (Yoshihama et al., 2005). We assessed landmarks, adverse childhood experiences (ACEs, 8 items, Felitti et al., 1998), community and school violence (adapted Things I Have Seen and Heard Scale, 11 items, Richters & Martinez, 1990), witnessing IPV (Revised Conflict Tactics Scale [CTS2] physical assault sub-scale, 11 items, Straus et al., 1996), physical maltreatment (CTS2 physical assault sub-scale, 11 items, Straus et al., 1996, plus severe spanking/whupping), and sexual victimization (rape and attempted rape via force by a non-partner, 2 items, from the National Intimate Partner and Sexual Violence Survey [NISVS], Black et al., 2011, plus forced touching), assessed yearly, beginning at age five. Additionally, we gathered demographic information such as participants’ primary caregivers’ highest grade completed (an indicator of socioeconomic status, or SES, Diemer et al., 2013), pregnancy and childbirth, and homelessness.

Starting with participants’ first partner (P1), we assessed for physical IPV (CTS2 physical sub-scale, 11 items, Straus et al., 1996), coercive control (adapted Brief Coercion Scale, 8 items, Cook & Goodman, 2006), and sexual IPV (rape/attempted rape via threats, force, or incapacitation, 6 items, from the NISVS, Black et al., 2011) in each relationship. We captured relationship characteristics such as its length and level of fear (0–10, with 10 = extremely afraid). For the current study, we defined severe physical IPV as being kicked, choked, hit, punched, beaten up, and/or assaulted with a weapon by P1, and we defined severe coercive control as any threats to hurt or kill her or someone she cared about by P1; since we operationalized sexual IPV as rape or attempted rape, any sexual IPV committed by P1 was defined as severe.

For the qualitative component of the larger study, we included a semi-structured interview at the end of each LHC interview for the first 40 participants (17 university, 12 two-year college, and 11 community). We conducted these additional qualitative interviews until we had reached adequate conceptual coverage of the core elements of IPV we were interested in exploring (i.e., saturation), across participants from each of the settings (Morse, 2007). With these qualitative interviews, we captured their in-depth experiences with different types of IPV and disclosure. The interviews were audiorecorded for transcription and analysis. For the current study, we analyzed the qualitative interview data from the 21 participants (10 university, 4 two-year college, and 7 community participants) who had experienced IPV during their relationship with their first boyfriend or partner, P1; additionally, we drew on their quantitative data for descriptive sample characteristics, e.g., their SES, prior victimization, and ACEs.

Of the 21 participants in the current study, 8 identified as Black or African American, 7 as White, 3 as biracial or multi-ethnic, 2 as Asian, and 1 as Latina. Participants ranged in age from 18 to 24 at the time of the interview (M = 20.67, SD = 2.15). Over half had received some sort of means-tested public assistance during childhood, and only one participant reported no ACEs (see Table 1 for sample characteristics). Over three-quarters of the sample had experienced community and school violence, witnessed IPV within their family, and been physically maltreated, with over one-third reporting sexual victimization by a non-partner. Participants’ age at the beginning of the relationship was just under 15 years old. Almost all reported physical IPV and coercive control from P1 (with roughly half of both types severe), while half reported sexual IPV (all of it severe). The physical IPV disclosure rate was 68%, compared to 53% for coercive control and 22% for sexual IPV. Two participants were currently in a relationship with P1.

Table 1 Sample Characteristics (N = 21)
Full size table

Data Collection and Analysis

University participants were interviewed in a private office at the university, while two-year college and community participants were interviewed in a private room at a community-based organization. Participants were interviewed by KP, after she received training by AK. At the start of each interview, the consent process was reviewed, emphasizing that participation was voluntary and she could stop at any time, or skip any question, and still receive compensation. In keeping with standard US practice to ethically provide an incentive for participation in minimal risk studies such as ours, participants were compensated $50 (Grant & Sugarman, 2004). As KP conducted each interview, AK reviewed the audiorecordings to begin data analysis. The combined interviews (the quantitative LHC interview plus the qualitative interview) averaged 82.14 min (SD = 22.44), while the qualitative interviews were 24.05 min on average (SD = 9.09). Each participant chose a pseudonym for herself and was told to refer to her partners as P1, P2, and so on. The qualitative interviews involved open-ended questions about each abusive relationship participants had experienced, and related disclosure.

The qualitative data for the current study, drawing on the interviews with 21 participants who reported a first relationship that was abusive, were analyzed using a grounded theory approach that involved open, axial, and selective coding (Strauss & Corbin, 1990). AK and EM analyzed the data. As we moved through the analysis, AK kept detailed process and analytic notes (Strauss & Corbin, 1990) in order to track the analytic steps and effectively build theory. In keeping with Tracy’s (2010) criteria for excellence, we emphasized rigor, credibility, resonance, significance, and meaningful coherence as we progressed through the analytic steps. AK developed the 13 initial open codes based on the qualitative interview questions, a deep reading of the interview with our first participant, Monae, and discussion with EM; this preliminary understanding is a point of departure for the analysis (Charmaz, 1996). After creating these codes, AK and EM each independently coded Monae’s interview, then we met to discuss our coding decisions and compute our inter-rater agreement; we sought to achieve at least 80% agreement (Miles et al., 2014). We achieved 70%, which prompted us to clarify a few codes. We each re-coded the first interview, and this time achieved 94% agreement. We then each separately coded 10 more interviews, dividing them evenly by setting. Once this open coding was completed, AK created a preliminary data matrix, and then EM drew on individual interviews to add summarized data to each cell. Displaying the data in this way allowed us to assess basic coverage across each code, begin to develop categories, their properties and dimensions (e.g., disclosure: how, when, which IPV type, to whom), and analyze initial patterns within and across participants (Miles et al., 2014; Strauss & Corbin, 1990).

Building off of this foundation, AK began axial coding, which involves developing connections across categories, using the paradigm model: Causal conditions➔ phenomenon➔ context➔ intervening conditions➔ actions/strategies➔ consequences (Strauss & Corbin, 1990). Focusing on the disclosure process as the phenomenon, we drew on the Kennedy et al. (2012a) model and participants’ data to build our paradigm. Though a grounded theory approach is not suitable for testing theory, per se, it is appropriate to draw on prior conceptualizations of a phenomenon in order to further develop connections between variables, extend our concepts, and deepen our understanding (Strauss & Corbin, 1990). Thus, our causal conditions included the three types of IPV and their characteristics (e.g., co-occurrence, severity, onset, chronicity, and fear)➔ which precipitated the phenomenon of the process of disclosure and non-disclosure of different types of IPV over time➔ as shaped by contextual factors such as the way she defined the problem (e.g., labeling her experience as abusive or not, minimization), the pros and cons of disclosure, stigmatization such as self-blame, shame, and anticipatory stigma, and the fit of her perceived needs with what was available and seen as helpful➔ and by intervening conditions such as social location; prior victimization; and community, developmental, and situational factors➔ which in turn influenced and resulted in her actions/strategies managing disclosure and non-disclosure, such as who did she tell, when, what, how, and under what circumstances➔ and finally, the consequences of disclosure, such as recipients’ reactions and degree of helpfulness.

Development of the paradigm led to another round of axial coding and matrix building, as we sought to further the connections between categories and their sub-categories: AK developed six new categories for each participant that reflected the paradigm and were derived from the qualitative data and some of the demographic quantitative data, as described earlier: 1) the different types of IPV and their characteristics, 2) her social location (e.g., race/ethnicity, public assistance and ACEs, homelessness), 3) her prior victimization and violence exposure, 4) her developmental context, 5) her situational context (e.g., her and P1’s ages, age difference, pregnancy or baby, did she live with P1, community or neighborhood factors, her and his family and peers), and 6) her disclosure process (including contextual factors, actions/strategies, and consequences of disclosure or non-disclosure). This process resulted in a second data matrix, which helped us deepen our understanding of the connections across categories, and set the stage for selective coding. Selective coding proceeds in relation to axial coding, and involves the synthesis of key categories into a coherent storyline (Strauss & Corbin, 1990). For the current study, the storyline can be understood as the role of multiple intersecting factors (social location; prior victimization and adversity; community, situational, and developmental factors; and stigma and stigmatization) in shaping young women’s process of IPV disclosure.

Because each young woman’s experience was unique, and because a key aspect of building theory is to develop understanding both of these unique aspects as well as the primary patterns across participants’ experiences (Bryant & Charmaz, 2007), the next step was for AK to derive core patterns that addressed our broad research questions, illuminated the storyline, and made novel, useful contributions to theory. The four theoretical patterns that emerged are: 1) factors that shaped the unfolding of intimate terrorism within these relationships (Johnson, 2006), key characteristics—including a quasi-parent-child dynamic between participants and partners—and how these factors and characteristics shaped disclosure, 2) the curious case of participants who experienced severe IPV but very low or no fear, minimization of the abuse, and no or limited disclosure, 3) factors that distinguished participants who sought help from LE, and their experiences with LE, and 4) the role of stigmatization, in concert with situational factors, in shaping disclosure. After completing this step, AK detailed her analysis in a memo for EM, which they discussed, and EM read through the interviews and second data matrix, to serve as a reliability check, paying particular attention to any disconfirming evidence (Miles et al., 2014). Because the goal of a grounded theory approach is developing theory by establishing conceptual connections between existing knowledge and new discoveries about a phenomenon (Strauss & Corbin, 1990), we present our findings in concert with our discussion.

Results and Discussion

Pattern One: Intimate Terrorism, a Quasi-Parent-Child Dynamic, and Disclosure

Four participants, Maya, Monae, Shakira, and Tiffany, lived with P1 and experienced severe physical violence and coercive control, a form of IPV termed “intimate terrorism” (Johnson, 2006); Monae was also raped. Intimate terrorism has not been studied widely among adolescents, but among adults it has been linked to injury, chronicity, and increased severity over time (Johnson, 2006). Ranging in age from 12 to 16 when the relationship began (with P1s 2–8 years older), these young women had many commonalities: All were poor, Black, and unstably housed prior to moving in with P1. They all witnessed and were victimized by extreme levels of community violence growing up; Monae, Shakira, and Tiffany, recruited from community sites, also reported severe and chronic witnessing IPV and physical maltreatment within their families. Only Shakira had not been sexually abused, while only Maya, a two-year college student, reported no child welfare or juvenile justice system contact. All four became pregnant with P1.

The combination of their disadvantaged social location, prior cumulative victimization, lack of stable housing, and very young age contributed to both their decision to move in with P1 and the longevity of the relationships, which lasted from 2 to 7 years. Tiffany was a 14 year old runaway when she first started living with her 16 year old partner, hidden in the attic of a house owned by his grandmother. Her mother was in prison and her father was in a gang. While they lived together, her partner controlled her access to food and her movements, and increased his physical violence and control during her pregnancy. She reflected on their relationship:

I wasted six years, and a lot of that was my child’s life, in that relationship. And I think it was because I was a runaway. That was the one thing he did for me. I didn’t really have to worry about where I laid my head at…at 14, when you don’t have nobody and nowhere else to go, it don’t sound so bad to keep quiet for that…It wasn’t just anybody that gonna go with a 14 year old runaway. You know? Maybe that’s why. Maybe that’s what it was. To not be lonely.

Shakira was 12 when she began seeing her 18 year old partner. At 13 she was pregnant, and her mother forced her to have an abortion and kicked her out of the house. She briefly lived with her grandmother, and then with P1; their relationship lasted 7 years, with onset of coercive control and physical IPV at 2 years. Shakira said, “Maybe if I would have had the baby I probably wouldn’t have stayed around so long. But, yeah…I think the only thing that played into it was the fact that he was taking care of me. He knew I didn’t have anybody else.” The younger a girl is when she begins a “statutory” sexual relationship, the older her partner is (Hines & Finkelhor, 2007, p. 300). Poverty and a poor parental bond influence these relationships, which can exacerbate girls’ pre-existing vulnerabilities (Wood et al., 2010; Young & d’Arcy, 2005).

Once they were living together and isolated from friends and family members, their partners assumed total or near total control. Stark (2007) captures how intimate terrorist partners subject women to “lessons” designed to teach them right and wrong, and shame them by demanding obedience to rules that might be used to discipline a child. These tools of coercive control took on additional meaning in the participants’ lives: They each noted ways in which their intimate partnership took on a parent-child dynamic. For example, Shakira recalled that her partner took care of her basic needs and helped her get to school each day; she described feeling “rebellious” when he was controlling or abusive. Monae, who was 16 and fleeing a sexually and physically abusive uncle (her legal guardian) when she moved in with her 24 year old partner, detailed his controlling violence, and how she coped as an abused child might:

I remember the first time he put his hands on me I wanted to run away…And he wouldn’t let me. When he goes to work, he’ll lock outside of the door... The windows would be sealed, couldn’t get out no kind of way. He’ll take my phone and he’ll take the house phone. And then make it to seem like if I ever were to scream, the neighbors wouldn’t hear me…I used to cry a lot…and then I—for some reason he called it the “go try to make yourself happy closet” and…after he would get done hitting me and the things he did, I would go in the closet and cry…I felt like that closet was the only thing keeping me away from getting hit.

One time, she tried to fight back, and he taught her a lesson:

He had slapped me and then told me that I never had the balls to hit him back. So I finally got a bat, and it did hit him, but it didn’t hurt him, it hit him on his knee. But then he took it out of my hands and hit me with it…And it kind of, ugh, gosh, “I will never do it again.” He had told me to learn not to hit people back.

Maya, who was 16 when she started seeing her 19 year old partner, and 18 and homeless when she first moved in with him, made the parent-child dynamic explicit:

I kind of feel like in our relationship that I was the daughter and he was the dad, almost. And I used to be like, “I only have one father. You’re not my father. I have already been raised up. I am raising my own kid, your kids. You’re not going to treat me like a kid.” And that was my biggest problem with him. I felt like he thought he was my father. He was very, very, very controlling.

She reacted defiantly: “If you want me to respect you, then you gonna respect me. So I think that created a lot of fights too because I was defiant. I kind of felt like, I don’t know, a slave that had a master and you wasn’t trying to do what the master said, so you keep getting the lashes.”

All four labeled their experiences as abusive, using such words as “beatings,” “almost killed me,” and “torture,” and all were extremely afraid, with fear levels of 9 or 10/10. Labeling, severity, and fear are positively associated with disclosure (Kennedy et al., under review; Littleton et al., 2006; Sabina & Ho, 2014), but their disclosure process was no simple matter, in part because of their disadvantaged social location and lack of informal support, but also because of the nature of the coercive control, which isolated them and limited their freedom (Øverlien et al., 2019; Stark, 2007). Both Shakira and Tiffany sought help from LE, which we discuss in the third section, so here we focus on the process of disclosure for Maya and Monae. Maya did not witness IPV nor was she physically maltreated in her family growing up, which made her reject disclosing to anyone close to her: “[I was] just embarrassed. I wasn’t brought up like that. So, to sit and talk to a family member or a friend that knew how I was brought up, knew how my family was, it just wasn’t something that I wanted to talk about. So, I just kept it to myself.” After 7 years of severe abuse and isolation, Maya reached out in desperation to her partner’s mother:

And she was like, “I tip my hat to you ‘cause you done been through a whole lot with him and you still keep trying to hold on…” And her words kind of comforted me because at least somebody sees all the sacrifice... She was like, “You know what, you don’t deserve that. Just leave him.” And those words just stuck in my head. If his mother is telling me to leave him alone, it’s time to go. So that was probably four months before I left...You are strong. You don’t need him.

This affirmation was the supportive encouragement she needed (Sylaska & Edwards, 2014). She left P1 without advance warning and traveled to a new city, where she attained help from a YWCA and domestic violence center; she also began taking classes at a two-year college. She felt “real happy” with the formal help, and “free” to enjoy a quiet family life with her daughter: “Just me being by myself and happy with my daughter. We don’t have bad days no more.”

Monae had no supportive family members or friends, except for her sister. One year into the relationship, after P1 had raped her, she fled to her sister’s and disclosed the rape. He came after her, broke into her sister’s house, and forced her back with him. She began plotting her escape, and was aided by key witnesses, who can facilitate disclosure (Black et al., 2008):

He did not want me to leave the house ‘cause he was afraid I was gonna go tell. Which that was my whole plan, but I told him I was going to the store, which we were, to get diapers and stuff…And I turned around and he was behind me, and I get punched…And he ended up punching my tooth out. And, I was in shock. I screamed. He was like, “Oh, you can scream all you want, there’s nobody to help you.”…But on that day, when that happened, a lot of people seen it. And, he got mad at me ‘cause everybody changed their direction toward him…[And] this lady came over and she had showed him how he was in the paper punching my tooth out. His address was on there and everything. So I knew, yes, it’s time…It felt like when everybody noticed that, I was free to go.

Once the physical abuse was public and his hold was broken, with the aid of the neighbor and her sister, who helped her pack and escape the apartment, she was able to leave him for good.

Pattern Two: Severe IPV, Low or no Fear, Minimization, and Limited or no Disclosure

Four participants, Angel and Alice (recruited from the university) and Casey and Jessica (recruited from the community) experienced severe IPV, yet reported very low or no fear. This finding is curious, as researchers have demonstrated a link between greater IPV severity and fear (Capaldi & Owen, 2001; Hamberger & Guse, 2002; Ross, 2012), with severity and fear associated with labeling and disclosure, as we saw with participants in the previous section (Kennedy et al., under review; Littleton et al., 2006; Sabina & Ho, 2014). By examining prior victimization and adversity; situational aspects such as family and peer influences (e.g., peer group maintenance of compulsory heterosexuality, Tolman et al., 2003), and IPV type; and how participants define and label their abusive experiences, we can begin to gain a theoretical understanding of how these factors interact to impede disclosure of severe IPV.

Angel, a 19 year old Black university student, reported that growing up, her mother was addicted to drugs and alcohol and had serious mental health problems. Throughout her early childhood, they were unstably housed and lived with “random other adults;” she lived with her father from age eight until college. Angel witnessed severe, chronic community and family violence, and experienced repeated physical maltreatment and sexual victimization over many years. When she was 15, she started dating P1, who was 19 at the time. On homecoming night, they attended a party, Angel drank heavily and passed out, and two of P1’s friends raped her, with P1’s assent. Angel reported zero fear and ambivalence about labelling it as rape, instead minimizing it as “two guys having sex with me.” She expressed fatalism about what happened:

It’s nothing you can do. I really, cause I don’t, I wouldn’t want to (pause) press charges against them in a sense. ‘Cause I don’t believe they were being criminally, as trying to really do that. I mean, they did know that I was unconscious. Which is still wrong…I just really wanted it to be not a thing. For it to be over…I wasn’t [afraid] because I just been in certain sexual situations before that I just expected—well, it’s not that I expected it—but I just accepted that these types of things happen to me. And just, I just felt like nothing good could really happen, especially from a partner.

Angel did disclose to a friend much later because she felt she had to, but did not tell her father, nor seek help from LE or anyone else. Her prior cumulative victimization and adversity, parents’ failure to protect her from abuse (Smith & Cook, 2008), and perhaps symptoms of post-traumatic stress disorder (e.g., dissociation and disengagement coping, Iverson et al., 2013; Kennedy et al., 2014), worked in tandem to cause her to minimize her abuse and limit her disclosure.

Casey, a 23 year old Black community participant, witnessed someone stabbed in the community, witnessed severe, chronic IPV in the family, was severely physically maltreated, and experienced sexual victimization by multiple perpetrators. She noted that her mother had serious mental health issues as well as a gambling problem, and that they were poor. Casey began dating her P1 at age 13; he was also 13, they had mutual friends, and he was “one of the popular kids.” He physically abused and controlled her, and attempted or forced her to perform oral sex multiple times. Casey reported low fear of P1 (2/10) and expressed ambivalence about what happened. Of the physical abuse, she characterized it as “…just a couple of arguments. It wasn’t really anything serious, serious where I was really scared.” She labeled the sexual IPV as “forced” but then minimized his behavior: “…he was 13, a horny little boy. I guess he really wanted it at that time…I wasn’t really scared or anything. I was more so nervous ‘cause I had never done that before…I also kind of felt bad. I didn’t want him to break up with me either.” In the context of her extremely severe, prior cumulative victimization, the abuse from P1 was not that serious (Few & Rosen, 2005). Illustrating one of the tenets of compulsory heterosexuality, she justified his sexual aggression as natural (Tolman et al., 2003) and did not want to jeopardize their relationship, particularly given the peer context and his high status (Barter et al., 2009). As a result, she did not disclose to anyone, even when talking with her friends about their abuse.

Sexual IPV may be especially likely to be minimized (Kennedy & Prock, 2018), as Alice’s story illustrates. Alice, a 19 year old Chinese American university student, reported witnessing a few fights at school, but no witnessing of IPV within her family. She was paddled by her mother from ages 5–7, and states that her mother was unemployed and they received food stamps when she was in early adolescence. She began seeing P1 when they were both 16; the relationship lasted 3 years. Alice described the sexual violence she repeatedly endured:

…towards the end it was pretty much rape every time. Because it was, it would just be like, he’d start touching me and stuff and I’d be like “Stop.” Wouldn’t stop. And then he would, we’d be on the couch and he’d literally pick me up and then take me to his bedroom and I’d be like, like this on the door frame (acts out pulling on the door frame) “Stop.” And “I don’t want to do this.” And he’d throw me on the bed and take off all my clothes. And then do whatever. And the entire time I’d be like “Stop, no.” And then, [he] wouldn’t.

In part because of her lack of experience, her idealization of P1 (Toscano 2007), and the norms of compulsory heterosexuality (Tolman et al., 2003), she understood forced sex by PI to simply be an integral aspect of the girlfriend role, and reported zero fear:

It sounds so generic, like, oh, clueless woman, but at that point, I really was just like, well if I’m in a relationship with a guy for this long, and if he wants it [sex] and I don’t really want it, it doesn’t really matter. I’ll just do it. It’s whatever…I

thought that’s just how it was. I thought that’s just, I was legitimately gonna marry this dude. You know? But, I mean, that was just prematurity thinking.

Her understanding of what had happened evolved over time: “…it wasn’t, it didn’t really occur to me that that was rape until after we broke up, a few years later.” In part because of these ambiguities related to sexual IPV, delayed labeling and acknowledgment is common, with lack of acknowledgment tied to reduced disclosure (Harned, 2005; Littleton et al., 2006, 2008).

Finally, Jessica’s story offers us a different look at the critical role of prior victimization and adversity, as well as family influences, in contributing to IPV experiences, low fear, minimization, and limited disclosure. Jessica, a 23 year old Latina recruited from the community, had an unstable childhood characterized by severe community violence, severe and chronic family violence and adversity (e.g., family members using drugs, depressed or mentally ill, attempting suicide, and in jail or prison), and severe, chronic physical maltreatment by her mother beginning in adolescence. She reports that she and her mother continue to have physical fights at present. Her family was poor, and she lived with her grandmother from ages 8–10. After her father died when she was 14 and her mother remarried, the family violence only worsened. After her father’s death, her older brother—who was himself very violent and in prison at the time of the interview—normalized the connection between family loyalty and violence:

When I was growing up it was always, there was six of us…it was if you see somebody doing something, or you see somebody fighting somebody, you have to help, no matter anything. That’s just, you know, once my dad died, it was more of my brother. That was his mindset. And if you didn’t, and you would come home, you would get in trouble for it.

For young women growing up in violent communities and families, this socialization into their own use of violence can make practical sense: You are either a loyal family member and fighter, or a victim (Kennedy et al., 2010). Jessica reported juvenile justice system involvement at age 15, and began dating P1 at 16, when he was 19. Their relationship was physically violent:

The first time was when I had went to his house but I had some of my friends in the car. And he accused me of doing things that I wasn’t doing. And he, you know, threw a chair…and it just escalated from there…I got out of the car and that’s when it got physical. We started arguing. He shoved my face. I hit him back. And then it was just full on from there.

Given her socialization into the use of violence, responding in kind was “normal” and she had low fear (3/10). How she labeled the IPV reflected this: “It got physical,” “we started arguing,” and “it was just full on.” In describing her use of violence, she used passive voice and connected it with her family: “I really try to stay away from it, but when it happens, it just happens…that’s how I was raised. I was raised a certain type of way.” Her normalization and minimization of the IPV, her need to avoid being seen as a victim (Wood et al., 2010), and her brother’s violent ways—and perhaps the reputation of P1—led her to reject disclosing to any family or friends: “I know if I would have told somebody and somebody would have reacted the first time, if would have just got horrible…Just the person that he was [P1], it was just that harmful.”

Pattern Three: Factors that Promote Seeking Help from LE, and its Helpfulness

Out of the 21 participants, three sought help from law enforcement (LE) related to their physical IPV victimization: Jane, Shakira, and Tiffany, the latter two of whom were introduced in the first section. This low rate (14%) of disclosure and seeking help from LE reflects the empirical literature. Based on the small number of studies available, young IPV survivors, whether adolescent or emerging adult, rarely go to LE for help (Sabina & Ho, 2014; Thompson et al., 2007; Watson et al., 2001). All three young women were from the community sub-sample and Black. Shakira and Tiffany were poor, while Jane’s family was middle class and her mother worked for the Department of Corrections (DOC). Each experienced severe, ongoing physical IPV and/or coercive control within their relationship, but no sexual IPV; they all reported high fear of their partner (8–10 out of 10). Finally, all three reported disclosing to informal supports who were close to them, such as a friend, family member, or new intimate partner.

This profile mirrors the available research on predictors of disclosure to LE by adult survivors (studies of factors associated with adolescent IPV disclosure to LE have not been conducted, to our knowledge): Greater severity (e.g., injury or perceived lethality) and disclosing to informal supports have been found to predict physical IPV disclosure to LE, while having a college education and being White predicted reduced disclosure (Chen & Ullman, 2014). Sexual IPV is less likely than either physical IPV or sexual assault by a non-partner to be reported to LE, in part because of shame and stigma (Kennedy & Prock, 2018; Weiss, 2010). There are mixed results on the helpfulness of LE by adult IPV survivors, with some women reporting that LE was unresponsive and unhelpful, others rating them as very helpful (Gover et al., 2013).

For Tiffany, Shakira, and Jane, there was a particular catalyst that prompted them to seek help from LE. In Tiffany’s case, near the end of her abusive first relationship with P1, she began a relationship with a new partner, whom she later married. She trusted him, and disclosed P1’s controlling, abusive behavior. Through this new relationship, she was able to begin extricating herself from P1, but he was unwilling to separate, forcing Tiffany to seek help from LE:

And what really made me stop talking to him was when my husband came along and we started hanging out as friends…That’s what took me away from that situation…They didn’t stay too far from each other. So he [P1] used to walk down there and he refused to get off the front of my car…First he laid in front of my car and said I couldn’t drive off without talking to him. I’m crazy. I’d of ran you over, you know? So I playing with the gas and the brake, and he get up and he hopped on my car. And he telling me, he hitting my window, “You need to get out and talk to me.” And I had to take that man—couldn’t get out of my car, ‘cause we was gonna fight—so I kept the doors locked and took that man to the police station on the hood of my car.

LE threatened to get child protection involved: “They said, ‘What is this?’ Just laughed at me and said they was gonna take my baby if we don’t get that together. If they see us again looking like that, they would take my daughter.” These sorts of responses are obviously unhelpful.

Over the course of their 7-year relationship, Shakira’s P1 began using cocaine heavily, and his physical abuse increased in severity. Her friends witnessed his violence and eventually stopped wanting to spend time with her, which he reinforced through his controlling behavior. Growing up, she had witnessed her stepfather physically abuse her mother; she was ashamed about this shared experience with her mother, and would minimize P1’s abuse. However, as his physical abuse escalated, she felt she had no choice but to involve LE (Chen & Ullman, 2014):

…I would never admit that it was as worse as it was because I didn’t want to be like my mom…But the beatings started getting worse. I mean, it was to the point where I would have to call the police every single time because once he started, he does not stop. And then if I ever do get the upper hand, then we would fight for the rest of the night.

LE would never hold him for the IPV, but ultimately he spent time in jail and a halfway house for drug possession, enabling her to end their relationship. In this way, the criminal justice system took his drug use more seriously than his severe physical abuse of Shakira.

Jane was 16 when she began seeing P1, who was 19. She did not witness IPV growing up, though she was spanked or whupped during childhood. In contrast to Tiffany and Shakira, she never lived with P1 and their relationship was relatively brief, lasting 7 months. Beginning at around 6 months, he became physically abusive, grabbing her and throwing her phone when he suspected her of texting another man, and severely threatening and controlling:

I had lots of stuff going for me. Playing basketball. I had a good head on my shoulders and stuff. So, he always said, “I want a kid by you ‘cause I know you’ll take care of him.” And just, the wording of that sounded like, I would be on my own anyways. So I never chose to have a kid with him. And so, he was like, “If you ever have a kid by anybody else, I’m gonna kill you and the kid. You won’t live to see him.”

Jane disclosed these threats to her family, including her mother who worked for the DOC. Her response was to tell Jane to stop seeing him; Jane agreed, and they obtained a personal protection order (PPO). A few months later, they returned home and caught him in the act of stealing their possessions. He was later sentenced to prison. For Jane, the support from her family coupled with her mother’s ties to LE enabled her to end the relationship soon after the physical abuse and coercive control began, and receive crucial assistance in the form of the PPO. Emotional support plus tangible aid from family members are critically important, especially for young survivors, and can facilitate the connection to formal providers (Kennedy et al., 2012a; Wood et al., 2010).

Pattern Four: Stigmatization, Situational Factors, and Disclosure

Most participants (17/21, or 81%) reported stigmatization, conceptualized as internalized stigma related to IPV in the form of self-blame, shame, or anticipatory stigma (Kennedy & Prock, 2018). Here we focus on four participants, Kelsie, Kayla, Jasmine, and Jo, and examine how stigmatization, coupled with factors such as peers, family influences, IPV type, and relationship status, shaped their disclosure. Kelsie, a 19 year old White university student, started dating P1 when they were both 13. She was middle class growing up, though her mother was depressed throughout her childhood and adolescence, and her parents divorced when she was 11, after which they shared custody. Kelsie reports that P1 hit her twice, and was controlling; her fear was 6/10. The first time he hit her was in school; afterward, Kelsie felt ambivalent and confused about his actions, and acutely aware of the social context and potential for stigma:

…it kinda seemed like an accident…Him and his friend were fighting in the hallway, and I was trying to tell them to stop fighting ‘cause there was a teacher coming…and I tried to get in between them, and I still don’t know if it was really an accident ‘cause then he punched me instead of his friend. And he said it was an accident, but I mean, I wasn’t anywhere near his friend, so it doesn’t really look like an accident to other people who saw it in the hallway, but you know, everybody just called it an accident…I think he was just mad that I was trying to interfere with him and one of his friends…I don’t know if he just, didn’t want people to think that I was, you know, controlling him or something….I was really surprised ‘cause I didn’t, especially ‘cause I didn’t know if it was an accident or not. So, and then I was really embarrassed ‘cause a lot of people saw it in the hallway…And he was my first boyfriend so I didn’t want them to think we were in a bad relationship, you know? And I didn’t want my teachers to find out.

With hindsight, she was able to add nuance to her interpretation and view his hitting her as volitional and about his own gendered image management (Tolman et al., 2003). At the time, the embarrassment about being perceived as being in a bad relationship—and the anticipated stigma—seemed worse to her than actually being in one. Thus, she minimized it and tried to move on (Barter et al., 2009). When he hit her again, she blamed herself: “I was just so mad at that point ‘cause I had kind of convinced myself that it was an accident the first time and so I felt stupid, when I let it, you know, happen again?...I just felt bad about myself and I felt like it was my fault.” In response, Kelsie did not disclose the physical IPV until years later, to a few friends. As noted, self-blame and shame are key impediments to disclosure; young adolescents may be especially unlikely to talk about IPV (Bundock et al., 2018; Kennedy & Prock, 2018).

Kayla, a 19 year old Black university student, began dating P1 when they were both 16. She and her sisters witnessed her parents’ fighting during childhood, and they divorced when she was 12. Afterwards, her mother warned her about not being “weak-minded” and letting a man abuse her. As a family, they all tacitly agreed not to discuss her father’s abuse: “…[it’s] weird. We never talk about it. We just, let it go…And it still puzzles me sometimes but I know I will never find out.” When she and P1 got into a disagreement 5 months into their relationship and he “lashed out,” pushing, grabbing, slapping, and hitting her, she was “very terrified” (her fear was 9/10) and tried to placate him in order to safely navigate the situation. Later, she re-evaluated his earlier comments (e.g., he had “jokingly” mentioned “smacking” her), assessed her placating reaction through others’ eyes, and anticipated stigma should she disclose to anyone:

You know, I didn’t know what to do because I didn’t want him to kill me. So at the time I just tried to be very calm, like “okay, okay, okay,” ant then you know, try to play it off…I was thinking at first it would be my fault because I didn’t recognize the signs at first. And then, when I had, while it was going on, I was very calm and I was like, “okay, I’m sorry, I’m sorry.” I should, I feel like I never should have had to say I’m sorry. I should have been like, “no, I am leaving.” But, at the time, I was so terrified, so I figured people would think I was being stupid or something. You know, even though he, he was hitting me. You know? Just afraid of how people would receive it…After, the next day, I still was crying. I felt like it was my fault in a way…I was just like, wow. What could I have done to, you know, see that? ‘Cause my mom, she—I said I would never let a guy do that to me. So, it was just like, wow, this actually happened to me.

A close, open relationship with a parent can help facilitate disclosure of victimization and the attainment of effective help (Smith & Cook, 2008; Wood et al., 2010). For Kayla, her mother’s warning, her family’s unspoken rule about not talking about such matters, her fear of a blaming reaction, and her self-blame all conspired to prevent her from telling anyone what had happened. She ended the relationship with P1, changed schools, and did not disclose to anyone. Years later, she did tell a subsequent boyfriend; she states that he was supportive and helpful to her.

As we have noted, sexual IPV appears to be less likely to be disclosed than physical IPV, in part because of a greater sense of shame and anticipatory stigma (Kennedy & Prock, 2018; Thompson et al., 2007). Jasmine, a 22 year old Black community participant, illustrates this dynamic. Jasmine began dating P1 when she was 15 and he was 19; they started living together when she was 16. Their relationship was characterized by physical and sexual IPV and coercive control and lasted 6 years; her fear was 8/10. Jasmine identified as a fighter (she reported getting into fights with peers at least yearly beginning at age 11) and described the physical IPV as “physical fights”: “We used to get into it.” Three years into their relationship, one of her friends came over and saw them fighting, and thus the physical IPV was revealed. Research on IPV disclosure among adolescents indicates that witnesses are associated with disclosure (Black et al., 2008). In Jasmine’s case, her friend finding out was not helpful: “She was mad…And she said why have I never said anything? And why am I still with him letting him mistreat me and do me like that?...she got to running her mouth telling everybody.” A few more years into the relationship, when P1 raped her, she felt “nasty…used. I was hurt.” In describing what happened, she took care to justify her situational vulnerability, seemingly to ward off any blame or stigma:

He was drunk. I was drunk. He was high and once again, I wasn’t in the mood ‘cause we had already had sex earlier that day. And, I had to go to work the next morning. And he was—I don’t know. I guess he really wanted some sex. And, I went in and got in the shower again, which he was supposed to get in after me, but he came and got in with me. So I tried to get out of the shower, and he pulled me and I fell. ‘Cause, you know, the shower was kind of slippery. I fell, hit my head. I got up. Went to my room and locked the door. I tried to call my friend to come get me. She wouldn’t answer. So I laid there. He came to the door banging on the door…And wanted to get in. I wouldn’t let him in. He kicked the door in. And, I don’t know, just got on top of me. I got to screaming. And, I don’t know. I don’t really remember what happened after that…He was holding me down really hard. And I was already naked ‘cause I had just got out of the shower.

As a result of the rape, she became pregnant and had the child. She never told anyone, perhaps in part because of her negative experience with disclosing the prior physical IPV (Kennedy et al., 2012a), coupled with her shame at not being able to defend herself from the rape (Weiss, 2010).

Jo, a 19 year old White two-year college student, was one of only two participants who was currently dating P1. Jo experienced severe community violence exposure and victimization; after her parents’ divorce when she was five, she witnessed IPV between her mother and a series of boyfriends. She reported no physical maltreatment, and declined to answer questions about prior sexual abuse. She began dating P1 when they were both 16; they lived together for a time after she became pregnant, then broke up when he started seeing someone else. At the time of the interview they had recently reconciled and were planning to live together again. Jo experienced coercive control and mild physical IPV (P1 threw something that could hurt) and her fear level was 5/10. Jo did not minimize P1’s actions, describing them as “rages,” and “freaking out,” but over and over during the interview, she expressed self-blame, anger at herself, and shame:

I think his anger and his resentment towards me, and my willingness to look aside when he would mistreat me, I think it escalated into just a really, really bad, aggressive, awful situation. Which I guess is partly my fault, because I did allow him to speak to me in disrespectful ways. I think it just slowly escalated with what I let him get away with…And I think, I’m so angry at myself for letting it escalate that far…I’m really angry. I essentially turned into my mother, you know? I let him treat me any way that he wanted to, with full disrespect.

Given that she was in the process of reconciling with P1, perhaps accepting some blame was a way to rationalize his behavior and her continued commitment to the relationship, and to feel like she had some control. Jo had no friends—in part due to P1’s coercion—but she did disclose to a cousin, who was not helpful. Her delayed disclosure to her parents reflected her desire to maintain the relationship with P1, her family dynamics, and perhaps her shame and self-blame:

My dad and I never really talked about things. He is kind of a reclusive sort of person. So he knew that we had conflict within our relationship. He knew there were problems. I did not explain to him the full situation until after the fact. I didn’t tell him the way that I was being treated or the things that were happening in our relationship. My mom knew some of it, but she has a history of being in really abusive relationships as well. I don’t know. They knew, but they didn’t really know the whole extent of the situation.

Jo’s experience of waiting to fully disclose to family until after they had broken up mirrors what researchers have found: Exiting or considering exiting a relationship is associated with survivors’ labeling of the abuse and subsequent disclosure (Hammond & Calhoun, 2007; Harned, 2005).

Summary of Findings, Limitations, and Implications

Our qualitative exploration of these transitional, early IPV experiences reveals disclosure as a complex process that is multi-determined and highly managed, with evidence of partial (e.g., disclosing physical but not sexual IPV), limited (e.g., disclosing to a new partner but not friends, parents, or formal providers), and delayed disclosure (e.g., waiting years after the relationship ended to disclose) across participants. These findings add to our understanding of IPV disclosure as a complicated process, echoing what we know about youth’s disclosure of other forms of interpersonal violence (e.g., child sexual abuse, Lemaigre et al., 2017). For some participants, disclosure was the first step in leaving the relationship, thus interrupting the abuse. In keeping with the empirical literature, most disclosure was to informal supports (Bundock et al., 2018), and sexual IPV was more likely to be minimized and not disclosed (Kennedy et al., under review). When a few participants—after experiencing severe physical IPV and/or coercive control—did disclose to formal providers, the response was decidedly mixed.

Four notable and novel theoretical patterns emerged from the data, each highlighting different combinations of factors working in concert to influence the disclosure process (Kennedy et al., 2012a). Pattern 1 captures the intimate terrorism (Johnson, 2006) endured by four participants, as well as a quasi-parent-child dynamic between themselves and P1. These novel findings extend Johnson’s typology to adolescent survivors, and introduce this disturbing dynamic for further study. Our analysis reveals the critical role that these participants’ precarious social location, youth, prior victimization, family characteristics, and homelessness play in shaping their IPV victimization as well as their disclosure process. Forced by circumstances to live with their partners, the severe IPV they experienced inexorably followed the severe violence they faced as children (Kennedy et al., 2012b), illustrating life course stability (Elder, 1998). The control and isolation imposed by their partners, in particular, made disclosure very difficult, though they all eventually managed to disclose to at least one person and escape the relationship. While researchers are beginning to examine coercive control within adolescent and emerging adult relationships, our findings are unique in highlighting the role of social location, familial abandonment and violence, and homelessness in shaping early and middle adolescents’ experiences with intimate terrorism and a quasi-parent-child dynamic with their first partner.

Researchers have demonstrated survivors’ minimization of IPV as a key barrier to disclosure (Miller et al., 2010; Sullivan et al., 2010); Pattern 2 reveals why some young women might minimize the abusive behavior, and thus fail to disclose. For these four, though they experienced severe IPV (i.e., being raped and/or hit), they had little to no fear, did not label their experiences abusive, and reported limited to no disclosure. For two participants, prior extreme adversity, abuse, and violence seemed to numb them to the sexual and physical violence they encountered via P1. The IPV was not “really anything serious” because it was, in fact, less severe than their earlier abuse (Few & Rosen, 2005), and there is “nothing you can do” because no one had ever intervened to stop the violence. Further, sexual IPV is particularly freighted with self-blame and shame, acting as a further barrier to disclosing (Weiss, 2010). Given these factors, disclosure may seem futile and senseless. For the other two participants, different forms of socialization were critical: One young woman identified as a fighter who had been taught by her family that physical violence was expected, instrumentally useful, and a sign of self-respect (Kennedy et al., 2010), while the other—in keeping with compulsory heterosexuality—had been conditioned to believe that forced sex was part of the girlfriend role (Harned, 2005; Tolman et al., 2003). For these participants, being hit or raped by their first partner defied categorization as abuse because, based on what they had learned, they did not see themselves as a victim. Again, disclosure makes little sense from this perspective. With these findings, we advance understanding of why minimization might occur, and highlight the immediate role that family factors, including severe victimization, appear to play in shaping adolescents’ disclosure.

Very few young IPV survivors disclose to LE (Sabina & Ho, 2014; Thompson et al., 2007; Watson et al., 2001). Our sample reflected this, with three out of 21 (14%) turning to LE; we explored the facilitating factors in Pattern 3. All three were Black, from the community, and had experienced severe physical IPV and/or coercive control, but not sexual IPV. Each had at least one close friend or family member they could confide in. These characteristics parallel the literature on factors linked to adult survivors’ disclosure of physical IPV to LE, including severity, being non-White, and disclosure to informal supports (Chen & Ullman, 2014). Sexual IPV appears much less likely to be disclosed to LE, even when severe (Kennedy et al., under review; Weiss, 2010). For these participants, the combination of severe, life-threatening physical IPV or coercive control (rather than sexual IPV), high fear, and at least one source of informal support was enough to propel them to seek out LE. However, only one participant, whose mother worked in corrections, received a helpful response. The other two did not fare as well. When young, poor, Black survivors face severe partner violence, they may have few community-based options for support and material aid (Miller, 2008). If they turn to LE in desperate need of help and are instead ignored, laughed at, or threatened with child protective services, they may reasonably avoid seeking help from LE and perhaps other formal providers in the future (Kennedy et al., 2012a). Very few studies have examined IPV disclosure to LE by adolescents; our findings go beyond previous work to illuminate the role of IPV type, along with the severity of the abuse.

Pattern 4 captures the role of stigmatization such as self-blame, shame, and anticipatory stigma (Kennedy & Prock, 2018), in concert with situational factors such as family influences and the peer context, in shaping disclosure. While most (17/21, 81%) study participants reported stigmatization, we focused on the experiences of four young women. In general, their abuse mirrored the victimization experienced by female relatives, exacerbating participants’ shame and self-blame: They felt “stupid” and “at fault” for not recognizing the early signs of abuse, failing to heed their mother’s warnings, and “letting him get away with it.” Peers were sources of confusion, embarrassment, and betrayal, with participants attempting to protect their image and avoid looking like a victim (Wood et al., 2010), especially in relation to sexual IPV (Weiss, 2010). In response to these factors, disclosure was delayed until many years after the relationship had ended, limited to only a few people, and partial, such that physical IPV was revealed but not sexual violence, thus extending the literature and illustrating the complexity of disclosure.

Regarding limitations: Our sample was purposive, thus the generalizability of our results is unknown. We did not examine psychological abuse, and for most participants, their accounts were retrospective. Additionally, we were not able to conduct member checks or follow-up interviews with participants, perhaps hindering our understanding. Despite the limitations, our findings on the complex interplay of factors that facilitate and impede disclosure may be useful as we build knowledge on these early abusive relationships and design effective approaches that can reach all youth. To facilitate disclosure, prevention programs must begin early and focus on family violence, gender norms, and sexual IPV, given it appears the least likely to be labeled and disclosed. Practitioners working with young women should be educated about all types of IPV, and actively support disclosure and problem-solving, without judgment (Barter et al., 2009).