For those who experience intimate partner violence (IPV), injurious effects extend beyond primary (i.e., immediately/directly resulting from specific incidents) and secondary outcomes (i.e., resulting from ongoing abuse, but not directly tied to specific violent incidents) (Coker et al. 2002). Tertiary (i.e., resulting from others’ reactions to primary or secondary symptoms) negative effects of IPV can involve identity-attacks from others’ stigmatizing communication. Even in stigma contexts less severe than IPV, such societal labels and negative treatments have been found to exacerbate depression and anxiety susceptibility (Frable et al. 1998). Thus, coupled with the already-detrimental experiences they must endure, IPV victims’ experiences with, and management of, resulting stigma are essential for scholars and practitioners to address. This research focuses on the actual ways that theorized IPV stigma is both experienced and managed by victims.

Using findings from an original study, I illustrate the ways IPV victims’ biological sex (H1, RQ2), experienced abuse-types (H2), and relationship classifications (H3) predict felt-stigma and the different strategies (RQ1) used to manage it. To begin, I integrate literatures on communicative identity-formation and IPV, framed in this study as physical (i.e., bodily injury intent) and/or psychological (i.e., intent to emotionally or verbally hurt or control) abuse outcomes in the context of diverse victim-types.

Identity Formation and Social Management

Within a social constructionist perspective, human communication involves re-enactment of particular identity aspects until entrenched in self and other perceptions (Goffman 1959). Inseparable from specific situations, identities are also tied to group affiliations and structural forces. Ultimately, through this framework, identities and accompanying labels are social products. In the case of constructs such as “male” or “female,” “victim” i or “abuser,” people use group labels to direct their social communication. But overall identities also result from situational communication, which confirms and/or challenges each identity performance. Because people are identified socially by group affiliation and personally by distinctions from other group-types (Deaux 1993), comparisons are continually made to gauge “roles” in society. When people’s IPV experiences become known, their subsequent identity performances often fail to meet audiences’ expectations for appropriate roles expected of healthy, functioning adults (Hertzog and Rowley 2014). This failed performance can result in management tactics, victims’ attempts to reconcile expectation-communication differences or to face stigmatization (Goffman 1959, 1963).

Drawing from subsequently discussed existing theoretical works, I herein conceptualize stigma as any characteristic tied to socially evaluative interaction that potentially threatens a person’s identity, communication abilities, and/or overall well-being. To date, socially stigmatizing attributions of IPV victims have included: (a) perceived-causal (e.g., passivity, femininity, masochism, emotional dependency; Dobash and Dobash 1978; Kessler et al. 2001; Litman 2003) and (b) perceived-outcome (e.g., poor self-esteem and varying psychological diagnoses; Fergusson et al. 2005; Hines and Malley-Morrison 2001) personality or “character” flaws; (c) perceived-causal (e.g., weak, small, pregnant; Eckstein 2010; McFarlane et al. 2000) and (d) perceived-outcome (e.g., bruises, broken bodies, gynecological issues; Campbell et al. 2002) physical traits; (e) perceived-causal (e.g., race, socio-economic status, cultural beliefs; Tang et al. 2002) and (f) perceived-outcome (e.g., reclusive, ostracized; Loseke and Cahill 1984; Walker 2000) social dynamics; and (g) perceived-causal (e.g., conflict management style; Holtzworth-Munroe et al. 1998) relational communication.

Social in nature, stigma is determined by dominant societal messages (Mill et al. 2010), which incorporate damaging labels to individuals failing to uphold governing expectations. Three qualities of stigma are particularly relevant to IPV victims: potential, effects, and management of stigma.

Stigma Potential

A precondition for stigma occurrence is possession of a devalued personal or relational quality; it may not be stigmatizing yet, but would be damaging if known (i.e., discrediting) (Goffman 1963). Stigmatizing conditions can affect people even when the condition is not public (Brashers et al. 2002). Because stigmatization involves internal, individual perceptions (e.g., “I feel guilty.”) and/or larger societal labels (e.g., “I don’t feel bad, but society says I should.”), IPV victims may feel burdened to manage their stigmatized identity, even when their abuse is private (Sullivan et al. 2010).

For IPV victims, a double-bind exists similar to that experienced by other “hidden” groups; silence minimizes public stigmatization but prevents help, whereas disclosure threatens privacy (Brashers et al. 2002). The extent to which victims embrace – whether for personal coping or public stigma management – a vulnerable identity interacts with a society’s consensus regarding that trait as stigmatizing, which affects available structural resources (Goffman 1963). Stigma has the potential to be identity-located in both the victim and/or the labeler.

In addition to affecting victims’ own perceptions, assumptions of stigma possession (e.g., similar-other generalizations) can limit the applicability of stigma research and theorizing (Link and Phelan 2001). Individuals often reject identity labels and associations with similar others, so labels are accurate only insomuch as someone affiliates with them (Schur 1980). Thus, degree and type of stigma should be assessed on an individual basis.

Stigma Effects: Identity Outcomes

Stigma outcomes are typically negative (e.g., poor educational attainment, interpersonal relationship failures, lower income levels, psychological deficiencies, and diminished housing options, medical access, quality of care, and physical health) (Major and O’Brien 2005). However, perceptions and outcomes of particular stigmas (Frable et al. 1998; Link et al. 1991) are influenced by personal and relational features (Link and Phelan 2001).

Individual Factors

The personal characteristic of victims’ biological sex is central to debates surrounding IPV. It is difficult to address IPV without considering sex, gender, and power as they relate to identity practices of perpetrators and victims. In most cultures, genders (e.g., masculinity, femininity) are attributed to particular sex identities (e.g., male, female) and embodied in relational interactions. In the context of IPV, femaleness, femininity, and victimization have long been connected (Alhabib et al. 2010). Although at least 3.2 million yearly physical attacks from female partners are reported by men (Tjaden and Thoennes 2000), most understandings of victimization frame feminine women as the main recipients of abuse (Alhabib et al. 2010). Further, recent research reveals that identification with a gendered sex role (e.g., masculinity) may differentially influence the likelihood of victimization for men and women (Daigle and Mummert 2014). As a result, male victims may struggle when faced with IPV stigma stacked on top of the fact that they are viewed as “abnormal” victims (i.e., male) of such stigma (Eckstein 2009, 2010). It follows that potential IPV stigma, or the internalization of their norm-deviance, should vary based on victims’ sex:

  1. H1

    Males will report more internalization of IPV stigma than will females.

Stigma is contextually influenced by power (Link and Phelan 2001). Society (situation, time, and place) determines who and what is stigmatized. To be effective, power-exhibitions of control over others must conform not only to situational norms, but also must address the identities accepted by participants in that time and place (Schur 1980). As a result, the degree to and manner in which stigma is experienced is believed to be directly tied to power individuals hold in other spheres of their lives (Major and O’Brien 2005). Potential stigma should thus vary according to sex and relational factors related to IPV (e.g., DeHart et al. 2010).

Relational Factors

Noted by Johnson (2008), rather than researching nuance among violent relationship types, most studies compare violent to non-violent couples, drawing overarching generalizations dichotomized solely between the two contexts. However, that approach is of limited use to those actually researching and/or working with IPV victims, who tend to converge on theoretical distinctions (if not on labels) among IPV relationships; distinct (formal and informal) typologies are often classified by coercive components, types of abusive behaviors, and abuse patterns and frequencies. As one example, Johnson’s (1995) relational IPV typology included relationships subsequently labeled situational couple violence (SCV), resulting from escalated conflict, and intimate terrorism (IT), involving perpetrator-patterned coercive control.

Vital differences found among IPV relationship types indicate that using perpetrated control as a distinguishing classification factor (e.g., IT/SCV ii ) is essential to applied understandings of violence, stigma, and relationships (Anderson 2008). To illustrate, in IT relationships, violent incidents are believed to occur with higher frequency than in SCV relationships. SCV is more likely to be mutually perpetrated by men and women, whereas women are more commonly researched as IT victims (Langhinrichsen-Rohling 2010). SCV is more common in society than IT (Johnson 2008). Additionally, SCV, as measured in large-scale family conflict studies, is not always perceived as abusive – by scholars or laypersons – in the same sense as IT. This confusion is attributed to SCV’s lack of patterned coercive control (attributed to IT), an absence of which leads people to perceive SCV as “normative” communication (Johnson 2011; Straus and Gelles 1990). To additionally test theorized IPV relationship differences, I proposed varying stigma experienced by victims:

  1. H2a

    The more physical abuse victimsi report, the greater stigmatization they will report.

  2. H2b

    The more psychological abuse victims report, the greater stigmatization they will report.

  3. H3

    IT victims will report higher levels of internalized stigma than will SCV victims.

Stigma Management

Regardless of circumstance, people with potential stigmas (e.g., homelessness, HIV, cancer) control or manage their conditions in such a way as to reduce others’ perceptions of weakness, vulnerability, and unworthiness (Brashers et al. 2002). IPV victims facing potential stigma must negotiate multiple factors when their victimization is (willingly or not) revealed (Eckstein 2009). In many contexts, revelation of private information is largely dependent on perceived risks of doing so (Afifi and Steuber 2009). In IPV relationships, these risks arise not only from outside repercussions, but also inside the relationship itself. Thus, one way to manage IPV and its stigma is to not reveal victimization to others. Other specific strategies to avoid revealing a stigmatized status include withdrawal, disengagement from, or denial of situations in which their identity is vulnerable (Major and O’Brien 2005).

Beyond hiding their status, IPV victims’ options for managing stigma include social, as well as internal, strategies. Victims can (a) blame discriminatory practices to protect self-esteem, (b) internalize blame to exert situational control, (c) separate from threatening situations, (d) strive to excel in discriminating domains, (e) increase affiliation with others who are similarly stigmatized, or (f) challenge stereotypes to distance themselves from the stigmatized group (Major and O’Brien 2005). These strategies vary in both individuals’ attributions of blame for their condition and identification with the stigma identity. When dealing with concealable stigmas, individuals have been more likely to use particular strategies (e.g., problem- versus emotion-focused coping) if other aspects of their identity (e.g., activist, educator) are relevant in the encounter (Brashers et al. 2002). Despite research on stigma management strategies in other contexts, to date, the actual social practices of IPV victims have not been systematically examined. Therefore, this study concluded with an exploratory inquiry based on the following:

  1. RQ1

    What strategies do victims most frequently report using to manage IPV stigma?

  2. RQ2a

    What differences, if any, exist between men and women in reported use of IPV stigma management strategies?

  3. RQ2b

    What differences, if any, exist between IT versus SCV victims’ reported use of IPV stigma management strategies?

Methods

Participants and Sampling

To increase heterogeneity of abuse experiences represented, targeted sampling procedures for vulnerable populations (e.g., Watters and Biernacki 1989) were used. Specifically, implementing efficacy- and appropriateness-maximizing recruitment practices (see Kaplowitz et al. 2012), standardized research invitations (i.e., project description, inclusion parameters, survey direct access link, researcher contact) were sent to general interest (e.g., shopping, cars, music, exercise) and violence- and family-specific chat groups and forums, as well as direct emails, to domestic violence agencies and men’s groups (e.g., fathering organizations, men against rape) for potential nationwide listserv inclusion. All groups’ – private/closed and public/open – moderators were first contacted for permission. The approved invitation was posted by 34 (out of 200 contacted) group-agencies and in approximately 350 (out of 900 contacted) online forums. Across a 3-month availability period, for participants who chose to indicate where they accessed the study, 39 % were invited via email (e.g., social network member or organization listserv); 57 % found the posting online (e.g., message board, forum, chat group); and 4 % were notified through a local organization in which they participated. The survey was available online for 3 months. Reinforcing safety, eligible (i.e., English-speaking, U.S. citizen, adult) participants responded – via an SSL-encrypted site erasing IP address and location info and generating random identification – regarding a physically or psychologically abusive former heterosexual partner. Because abusers frequently monitor their victims’ behaviors, only people currently “out” of IPV relationships were recruited for this online study. Certainly, former partners may still stalk or harass victims, but focusing the study solely on “former partners” was an attempt to minimize current danger to those still in abusive situations – a method that also contributes to practitioners’ applied knowledge when counseling those recovering from IPV situations. Thus, the results of this study may be generalizable only to U.S. citizens self-identified as former victims.

Participants (239 women, 106 men; N = 345) were aged 18 to 72 years (M = 42.12, SD = 11.59), primarily White (87.4 %), and educated with some college (31.0 %) or a bachelor’s degree (28.7 %). Further information on participants’ demographics and relationships was detailed by Eckstein (2011). Ranging from 2 months to 55 years (M = 8.98 years, Mdn = 6.75 years, SD = 8.06), relationships ended an average of 7.13 years prior to the study (Mdn = 5 years, SD = 7.67, range = 1 week to 40 years), with abuse beginning M = 1.82 years (Mdn = 0.54 years, SD = 3.49, range = immediately to 30 years) into the relationship.

Procedures and Measures

By clicking the link in the invitation-posting, participants were routed directly to a page explaining the study in more detail; if interested, participants clicked again to access an informed consent page and were encouraged to print it for their own records (it also contained sources for IPV helplines and counseling); finally, participants opted to either leave the page or to click once more to indicate consent and to proceed to the survey. To respect a process of ongoing consent in a web collection process, attempts from participants completing less than 60 % of the entire survey (n = 108, including those not proceeding past the consent page) were deleted, as this was interpreted as remote-indication of desire to withdraw from the study.

For each measure, results from confirmatory factor analyses demonstrated the unidimensionality (i.e., face validity, internal consistency, parallel nature) (Gerbing & Anderson, 1988) of scale items. Specifically, fit parameters included: ratio of chi-square to degrees of freedom (CMIN/DF or χ 2/df) to evaluate model complexity adjustment (<3.0; Browne and Cudeck 1993), comparative fit index (CFI) to compare scale covariance to an established instrument (> .90, but scores closer to Hu and Bentler’s (1999) .95 were attained by most measures), and root mean square error of approximation (RMSEA) to determine model fit standardization (< .10; Byrne 2001). These specific results showed all final scales and subscales with satisfactory fit and are available from the author.

Intimate Partner Violence

IPV measures included scales of physical perpetration iii and victimization, psychological perpetrationiii and victimization, coercive control, and fear.

The Conflict Tactics Scales 2 (CTS2; Straus et al. 2003) and non-overlapping Partner Abuse Scale-Physical (PASPH; Hudson 1997) items assessed physical abuse. Participants rated overall relationship frequencies (0 = Never to 6 = Always) of their own and partner’s physical behaviors (e.g., Slapped around face or head, Beat up badly), resulting in nine-items measuring physical perpetration (M = 1.16, SD = 0.32, α = .78) and 12 items used to measure physical victimization (M = 2.36, SD = 1.25, α = .92).

A sex-neutral Index of Psychological Abuse (IPA; Sullivan and Bybee 1999) measured ridicule, harassment, criticism, and emotional withdrawal (e.g., Tried to humiliate, Discouraged contact with friends). Fifteen perpetration items (M = 1.67, SD = 0.65, α = .86) and 19 victimization items (M = 3.83, SD = 1.31, α = .91) rated relationship frequency (1 = Never to 7 = Always) of psychological abuse, with a nine-item subscale of domination and manipulation used to measure coercive-control (M = 3.78, SD = 1.34, α = .79).

Higher levels of fear are theorized for IT than for SCV victims (Johnson 2008). Thus, Peralta and Fleming’s (2003) five-item (e.g., Was afraid, Felt unsafe when with him/her) scale was used to evaluate victims’ fear felt (1 = Never to 7 = Always) during the relationship (M = 3.91, SD = 1.53, α = .91).

Finally, SCV and IT relationship types were differentiated in post hoc, hierarchical cluster analyses, an analysis method used to distinguish Johnson’s (1995) relational types in previous research (e.g., see Johnson 2008 for listing of studies; Leone et al. 2007). Mean score variable differences were used to sort participants into mutually exclusive groups and each case (i.e., participant score) was criterion-modeled into the best fitting “cluster” until only requisite clusters remained (Romesburg 1984). Following Johnson’s (2008) conceptualization of fundamental SCV/IT differences, victim-perpetration scores in this study were used only to distinguish IT from SCV relationships as a key relational factor in predicting IPV variance. Thus, a two-product cluster criteria included theorized differences on the six IPV variables: psychological and/or physical abuse perpetration and victimization, and coercive control tactics and fear received. Squared Euclidean distance scores found all cases with between-groups methods linking smallest mean differences significant at α < .05 (Romesburg 1984). Out of 345 participants, fewer overall (n = 126, 36.5 %) and fewer men (23.8 % of all ITs) than women (76.2 % of all ITs) experienced IT (n = 30 or 28.3 % of all males and n = 96 or 40.2 % of all females). SCV (n = 219, 63.5 %) was experienced by 71.7 % of all men (n = 76, 34.7 % of all SCVs) and 59.8 % of all women (n = 143, 65.3 % of all SCVs).

IPV Stigma

Due to similar characteristics (e.g., concealable with comparable management dynamics; see Mill et al. 2010), a modified HIV Stigma Scale (Berger et al. 2001) measured internalized IPV stigma. Participants rated (1 = Strongly disagree to 7 = Strongly agree) typical interactions with others on personalized stigma (e.g., Felt set apart, isolated from the rest of the world), disclosure concerns (e.g., Told people close to me to keep my abuse a secret), negative self-image (e.g., People’s attitudes made me feel worse about myself), and concern with public attitudes toward the stigma (e.g., Abused people are treated like outcasts). Mean scores (M = 3.59, SD = 1.42, α = .95) of 22 items represented IPV stigma.

IPV Stigma Management

Link et al.’s (1991) scale on managing mental illness stigma was content-modified to measure which strategies victims agreed (1 = Strongly disagree to 7 = Strongly agree) using to socially manage IPV stigma. Three subscales included: five items for Secrecy (e.g., Hide abuse experiences from others) (M = 3.19, SD = 1.30, α = .72); four items for Educating Others (e.g., Help people close to me understand what an abuse victim is like) (M = 4.78, SD = 1.47, α = .80); and five items for Situational Withdrawal (e.g., Try to avoid people with negative opinions about victims) (M = 3.93, SD = 1.31, α = .67).

Results

Use of cluster analyses to differentiate IPV relationship types is not without flaws (e.g., sample variability). Therefore, each analysis in this study was run separately using clustering and then interval scale data. Found significance did not differ by analysis method, so both clusters and continuous variables are reported when relevant; where redundant, only the most appropriate statistics are provided.

Preliminary Analyses

Physical and psychological victimization were positively correlated with one another (r = 0.62, p < .01) and with IPV stigma (physical: r = 0.25, p < .01; psychological: r = 0.34, p < .01). Independent samples t-tests showed women reporting both more physical (women: M = 2.48, SD = 1.32; men: M = 2.10, SD = 1.01) [t (258.53) = 2.87, p < .01, d = .36] and more psychological (women: M = 3.94, SD = 1.32; men: M = 3.59, SD = 1.27) [t (343) = 2.32, p < .05, d = .25] victimization than men. Men (M = 3.52, SD = 1.33) and women (M = 3.61, SD = 1.46) did not significantly differ in reported IPV stigma, t (334) = 0.53, ns.

Substantive Analyses iv

Abuse (H2) and IPV Stigma (H1, H3)

Predictions of sex differences in stigma were not supported (H1). Hypotheses predicting positive associations between IPV stigma and physical (H2a) and psychological (H2b) abuse were initially supported in preliminary analyses. H2 implied that these associations should persist for all victims, accounting for participants’ sex and IPV relationship type. To test this logic, two five-step hierarchical regression models (Model 1: physical abuse; Model 2: psychological abuse) were employed with IPV stigma as the dependent variable. In step one, a dummy coded variable represented sex (male = 0, female = 1). Next, IPV relationship type, determined via cluster analysis, was entered (SCV = 0, IT = 1). Step three included the independent variable (abuse type). Step four added two-way interaction terms (product of each of the first three variables multiplied by one another in pairs). Finally, models included a three-way product interaction term (see Table 1).

Table 1 Hierarchical regression of victims’ sex, IPV relationship typea, and abuseb predicting IPV stigma
Full size table

Hierarchical regression results indicated that sex did not predict IPV stigma (H1), but IPV relationship type (H3) (step two) did; IT victims reported more IPV stigma than SCV victims. On the third step for each model, after covarying sex and IPV relationship type, H1a and H1b were supported; physical (Model 1) and psychological (Model 2) abuse were positively associated with IPV stigma. None of the two-way interaction terms predicted IPV stigma (step four), but both models’ three-way interaction terms (step five) of sex, IPV relationship type, and abuse type explained additional IPV stigma variation.

To clarify the direction of these effects, interactions including dichotomous independent variables (sex, IPV relationship type) were decomposed, according to Aiken and West (1991). In hierarchical regression equations, the simple slope was run as a dependent variable (IPV stigma) regressed onto an independent variable (physical abuse) for different subsamples of participants: male SCV victims, male IT victims, female SCV victims, and female IT victims. These groupings of male versus female outcomes among different relationship types analyzed additional nuance in the initially unsupported H1. Each subsample was independently tested regressing IPV stigma onto IPV relationship type, physical or psychological abuse, and a two-way interaction term computed as the product of the two independent variables. v Unstandardized regression coefficients are reported to allow for conclusions drawn from small sample subgroups (i.e., 76 male, 143 female SCVs; 30 male, 96 female ITs).

For men, physical victimization did not predict IPV stigma for either SCV (B = 0.41, ns) or IT (B = −0.03, ns) victims. For women, physical victimization was positively associated with IPV stigma for IT (B = 0.34, p < .01) but not SCV (B = −0.17, ns) victims. For men, psychological victimization was positively associated with IPV stigma for SCV (B = 0.62, p < .01) but not IT (B = −0.57, ns) victims. For women, psychological abuse was positively associated with IPV stigma for both SCV (B = 0.28, p < .05) and IT (B = 0.42, p < .01) victims.

Finally, H3 predicted that IT victims would report more stigma than SCV victims. An independent samples t-test supported H3: IT victims (M = 4.05, SD = 1.46) reported more IPV stigma than did SCV victims (M = 3.32, SD = 1.33), t (334) = 4.71, p < .001, d = .52.

IPV Stigma Management (RQ1, RQ2)

RQ1 asked which communication strategies were most prevalent in managing IPV stigma. Paired samples t-tests comparing strategies showed Educating Others (M = 4.78, SD = 1.47) was used most frequently, followed by Situational Withdrawal (M = 3.92, SD = 1.31) and Secrecy (M = 3.19, SD = 1.30) (see Table 2). Educating Others was reported more than Situational Withdrawal [t (321) = 7.26, p < .001, pooled Cohen’s d = .62] and Secrecy [t (321) = 12.18, p < .001, pooled Cohen’s d = .15], and Situational Withdrawal also differed from reports of Secrecy, [t (322) = 11.82, p < .001, pooled Cohen’s d = .56].

Table 2 Mean differences between stigma management strategies by sex and IPV relationship
Full size table

RQ2 asked if IPV stigma management strategies differed by (a) sex and/or (b) IPV relationship type. Independent samples t-tests indicated no significant difference between men and women in use of strategies, and no differences between SCV and IT victims’ use of Secrecy and Educating Others. However, IT victims reported using Situational Withdrawal more often than did SCV victims (see Table 2).

Discussion

IPV stigma and its management differed among victims primarily by abuse type and IPV relationship type. As such, for each construct studied, there are theoretical implications for scholars. The concurrent practical applications for IPV victims and professionals are discussed in each topical section to follow.

Stigma: Implications for Scholars and Counselors of Identity and IPV

IPV relationship type and abuse type (H2a, H2b), rather than victims’ sex (H1 unsupported), predicted stigma. IT victims reported more stigma than SCV victims, lending credence to conceptual distinctions between IPV relationships distinguished by control and fear and differences in victims’ outcomes – both internal/personal and external/social in nature. Physical (H2a) and psychological (H2b) abuse also positively associated with stigma. These findings, with variable-interactions more complex than simple linear predictor trends, suggest that further nuance should be added to Johnson’s (2008) IT-SCV typology.

Physical IPV

Being female and experiencing an IT relationship resulted in higher stigma than for other groups. Because women are more likely than men to experience injurious, physical violence that is conspicuous (and also elicits negative reactions from society; Sullivan et al. 2010; Straus et al. 2003), this finding may be intuitive; injury necessitating medical care forces victims to reveal their situation. Previous research finds that female IT victims may be more likely than female SCV victims to seek formal social support from police officers, physicians, and counselors and to encounter complex issues and/or negative reactions when opting to seek assistance (Eckstein 2010; Johnson 2008; Leone et al. 2007; Sullivan et al. 2010). Although untested in this study, these factors may play a role for IT women involuntarily facing stigmatizing reactions, because even without professional treatment, it is possible for others to see physical signs of violence against women.

According to stigma theorizing, physically observable discreditable characteristics result in social embarrassment and/or identity invalidation (Goffman 1963). In the U.S., although at least 5.3 million physical attacks against women are reported each year (Tjaden and Thoennes 2000), victimization is still considered socially deviant and in many cases blame-worthy (Sullivan et al. 2010). In this study, female IT victims’ stigma increased along with the physical abuse they received. In our culture, men both have a diminished likelihood of experiencing visible IPV injuries (Straus et al. 2003) and/or encounter more acceptable social responses to visible bruises (e.g., physical labor, sports, brawls with other men); this may account for the finding that male IT victims’ physical victimization was not tied to their felt-stigma in this study. Further, the study’s comparatively small sample of men, some recruited from support groups, may have come (at the point of post-IPV status in this study) to a different understanding of their experiences, compared to what they actually experienced at the time of IPV. Thus, counselors working with this type of victim may wish to incorporate coping and/or abuse-management styles to address not only the IPV but also the stigma potentially associated with different styles. For example, scheduling support groups in public locations (even those devoted to safety and/or health services) or at inconvenient (or otherwise highly populated) times may limit resources for those not wishing to disclose their status to others in their lives.

SCV occurs more in the U.S. population than does IT (Straus and Gelles 1990), but SCV may entail less severe/sustained physical violence and does lack the fear/patterned coercion commonly found with IT (Langhinrichsen-Rohling 2010). This study’s finding that physical abuse was unrelated to stigma among male and female SCV victims supports stigma theorizing (e.g., Goffman 1963), as it suggests that people not frequently physically victimized (i.e., not managing external visibility) are less likely to report experiencing stigma. Thus, this study not only supports general stigma theorizing, but also provides nuance to IPV relationship typologies by adding stigma to other experiential and outcome differences between SCV/IT victims (H3).

Psychological IPV

This study also reinforces conceptualizations of stigma as an outcome of sometimes-hidden, unequal power relationships. Discreditable character stigmas – those for which people are considered blameworthy once they are discovered – are particularly stressful for people trying to concealing them (Frable et al. 1998). Current findings support the notion of psychological victimization (if hidden from others) as a discreditable stigma, which victims worry will be exposed (Sullivan et al. 2010). One explanation for this finding is victims’ personally-felt responsibility (as opposed to societal-culpability) for their IPV. Added psychological attacks from the perpetrator compound that stigma distress (DeHart et al. 2010). Therefore, the finding that female IT and SCV victims’ psychological abuse corresponded with increased stigma lends empirical evidence to theories of psychological IPV as a stigmatizing experience. Further, this finding suggests that female victims who experience psychological abuse – in a variety of forms – may encounter additional barriers to seeking support than do those experiencing largely physical abuse alone (possibly a moot point, as physical abuse is often accompanied by psychological control tactics). These potential, additional obstacles to identifying the psychologically victimized (and correspondingly offering tailored support to them) should be considered when support or informational campaigns reach out through public methods such as take-home flyers, handouts, or interpersonal interactions in any setting.

Unexpectedly, stigma was found to increase along with psychological abuse for male SCV victims but not for male IT victims. On the surface, this finding appears to contradict stigma theorizing regarding IPV victims; victims more severely abused should experience more stigma. However, probed closely, these data support Goffman’s (1963) hypotheses of stigma as varying across personalities, cultures, and situations. For example, maintaining an “appropriate” masculine identity while victimized is challenging (Daigle and Mummert 2014), because cultural alienation, denigration, and limitation of resources are largely sanctioned for abused men (Eckstein 2010). As categorized in this study, men who experienced SCV did not report psychological abuse as frequently as did male IT victims, who experienced more severe and patterned psychological abuse. In other words, since they were not used to psychological abuse, the comparative uncommonness, or identity incongruence, of the IPV experience for male SCV victims may have led to their feeling shame, embarrassment, and/or insecurity – factors evidenced in the stigma measure. In contrast, male IT victims may have become more accustomed to their psychological abuse and so may have seen their experience not as deviant but as a normal, expected part of their relationship (e.g., see Eckstein 2010 for male victims’ IPV rationalizations). Previous research suggests that, rather than affiliate with an embarrassed stigmatized victim identity, these men may instead view psychological abuse as a problem to solve; accustomed to habitual verbal and emotional attacks, they may justify their situation with masculine discourse related to protection, heroics, and stoicism (Eckstein 2009). It may be especially effective to use similar appeals to masculine discourse when offering information and/or support to these men. Ultimately, knowledge of male IPV victims’ experiences is limited, and Johnson’s (2008) SCV/IT conceptualization is rarely tested with male victims – a population that should be addressed in future IPV research.

Stigma Management: Practical Applications for Practitioners and Victims

Most IPV coping research focuses on victims’ management of physical or psychological abuse or on their stay/leave decisions (Sullivan et al. 2010). Compared to studies that focus on managing IPV experiences (i.e., as onset controllable cause), very little attention is given to victims’ management of IPV stigma (i.e., regulative empowerment outcome). This study provided an initial step in applying identity management strategies to IPV victims. Current results demonstrate that IPV victims’ stigma management strategies may parallel HIV stigma management, an area more commonly researched in the stigma literature (see Brashers et al. 2002; Mill et al. 2010). Thus, practitioners may want to consider the strategies victims in this study reported using to manage their identity experiences in former IPV relationships. Counselors may find it useful to recognize – and accordingly, distinctly treat – the different identity outcomes associated with an unhealthy relationship experience (i.e., IPV), a relational stigma (i.e., “failure as partner”), and a gender-stigma (i.e., “male victim”) – emphasizing personal stages of change, normative relational standards, and masculinity expectations, respectively.

Having implemented a retrospective research approach, current findings can aid therapists working with people trying to move on while managing multiple identities (e.g., “victim/survivor” and “stigmatized”). Having people report on past experiences, although they may re-frame the past, nonetheless allows for self-reflection which victims may not be prepared for (or capable of) in currently abusive contexts. Indeed, knowing how former-victims experience and manage stigma – during and after IPV relationships – can provide clinicians with practical information for helping victims at all stages of the stay-cope/leave-recover process.

The most common IPV stigma management strategies reported by victims were Educating Others, Situational Withdrawal, and Secrecy, respectively (RQ1). In previous IPV research, Educating Others was not a common tactic. For example, Eckstein (2009) found that IPV victims more often reported withdrawal, secrecy, and intrapersonal coping. Intuitively, severe victimization typically results in avoidance of negative societal reactions (Sullivan et al. 2010). Indeed, present stigma management strategies varied according to IPV relationship type (RQ2). SCV (with less severe/patterned physical and psychological victimization) victims reported using Situational Withdrawal much less than did IT (with more coercive control and fear experienced) victims. This finding corroborates IT as a relationship type causing victims to retreat, or disassociate, from dealing with stigmatization (Sullivan et al. 2010). Seeing these behaviors, therapists could differentiate treatment for Situational Withdrawal as a stigma management strategy (i.e., intentional identity coping tactic, albeit perhaps unhealthy) versus withdrawal as a psychological manifestation of injury (i.e., involuntary negative cognitive reaction).

One explanation for finding Educating Others as the most common strategy reported in this study could be participants’ overall formal education (79.6 % attended and/or completed college). Formal education and/or formal IPV-centered groups (from which some participants were recruited) may prepare victims to deal with stigma in sophisticated ways or may provide for them an alternatively empowered identity. If this is the case, primary prevention education strategies incorporated throughout school curriculum may not only be useful in fostering healthy relationship skills, but may then better equip those who do experience IPV to better identify the perpetrator (as opposed to recipient) as non-normative.

Another explanation for finding Educating Others as a frequent strategy aligns with research on secrecy. If diversity and directness of strategies for sharing sensitive information vary by individual factors (Afifi and Steuber 2009), it is possible that this sample – reporting on former relationships – had cognitively and emotionally progressed from “victim” to “survivor” identification. Differing internalized stigmas within this sample may have resulted in activist/educational stances toward coping, which Brashers et al. (2002) described as typical among those who focus on the problem, as opposed to their own personal emotions. Ultimately, people in abusive relationships may utilize different stigma management strategies than people out of IPV. In efforts to protect themselves and/or maintain their relationships, still-involved IPV victims do not always resort to Educating Others to manage their stigma (Eckstein 2009).

Finally, desires to educate others may not always be constructive – for self and/or others. Because stigma exists culturally and is reinforced through social interactions, people’s attempts to manage stigma by surmounting harmful labels may not address dominant norms, which can cause management strategies to fail (Major and O’Brien 2005). In addition to dealing with all the potential drawbacks of revealing a vulnerable identity, IPV victims must also monitor the messages, meta-messages, contexts, and environments of their revelation; co-manage their own and others’ potentially negative emotions; and adjust each aspect for every specific individual.

Considering all of this information can be useful to practitioners tailoring interventions to victims’ specific experiences. Counselors can use assessments of internalized social stigma to advise patients how (dependent on appropriate gender expectations for each victim), when (dependent on victim’s personal stage of change in IPV coping processes), and why (dependent on IT/SCV relationship type, corresponding to IPV frequency/severity) to reveal victimization. IPV victims could be encouraged to adopt stigma management strategies that are comfortable to enact, suitable to current coping abilities, and contextually appropriate for people in the process of leaving an IPV relationship.

Limitations and Future Directions

A primary limitation of this study may be a reliance on retrospective self-reports. Objective accuracy is compromised when victims evaluate relationships that ended, on average, 7.13 years prior. Whereas frequent behaviors may be accurately recalled only if patterned (e.g., once a week) (Menon 1994), irregularly occurring or severe IPV behaviors (e.g., beaten near daily) can cause participants to meld memories and may preclude discrimination of individual acts in short time spans. Although this study emphasized victims’ subjective perceptions, their reports remain important, as they may reflect coping mechanisms as currently constructed (and that seemingly “worked”) for victims.

An additional limitation to this study is the post-hoc assignment of participants to SCV/IT groupings. When grouping variables are also used as predictors (as were some of the IPV variables in this study), the ability to generalize about the independence of clustered groups can be limited. Further, this method presents a challenge for future replication using cluster techniques (even using pre-set criteria), as the groups themselves are within-sample comparative and so would differ according to population/sample. Thus, although theoretically informative, caution should guide application of results to specific individuals.

Counselors may be likely to first encounter abuse victims when they are at the same life stage as many people represented in this study. As a result, the disparities in stigma management strategies found here demonstrate the need for future research to explore associations among personal characteristics (e.g., psychological coping ability, relationship status, victim-identification), social characteristics (e.g., power position in society), and stigma management. Although findings from this sample may be limited in generalizability, it is possible that the stigma surrounding IPV differs based on perceptions of culpability. In other words, victim-, abuser-, and mutually violent-stigma may all comprise different social variables and thus management tactics. Research contributing to relational- or victim-interventions should clearly delineate IPV relationship types so that practitioners can tailor their counseling to help victims manage IPV stigma, cope with psychological outcomes, and choose effective communication strategies for self and others.

Conclusion

The associations found in this study – among abuse types, IPV relationship types, and IPV stigma and its management – illustrate the importance of IPV research in helping to shape a functional approach for people’s lives. Higher levels of stigma were found overall among victims of control-based relationships, when compared to those not founded on fear/control. However, when looking at the abuse that comprises the different relationships, a more nuanced picture emerged. Physical abuse predicted stigma only for female victims of intimate terrorism, whereas psychological abuse predicted stigma for all females and for male victims of situational couple violence. These results suggest that both public campaigns (e.g., aimed at reducing IPV stigma to promote help-seeking) and specific counseling approaches should be tailored to specific types of victimization, based on the presence/absence of varying levels and types of abuse. This research reinforces the importance of accounting for relational-, societal-, and personal-level variables when studying IPV.

Notes

  1. i.

    Recognizing labels’ power, I employ “victim” to emphasize experiences while subject to IPV; assigning a “victimized” identity is not intended.

  2. ii.

    Due to an existing publication emphasis (e.g., see Anderson 2008) on distinguishing patterned, controlling violence from other kinds of violent relationships, Johnson’s (2008) “types” of violent resistance and mutual violent control were not included in this study.

  3. iii.

    Perpetration scores, differentiating SCV/IT victims (e.g., retaliation) were not included in substantive analyses.

  4. iv.

    In two-tailed significance tests (α = .05) on data from 345 participants, statistical power for t-tests was .46 for small (d = .20) and .99 or higher for medium (d = .50) and large (d = .80) effects. Hierarchical regression power was .65 for small (ƒ 2 = .02) and exceeded .99 for both medium (ƒ 2 = .15) and large (ƒ 2 = .35) effect sizes (Faul et al. 2007).

  5. v.

    Per Aiken and West (1991), when coding IPV relationship type as SCV = 0 and IT = 1, a slope for physical or psychological IPV (two-way interaction step) denotes the association of physical or psychological abuse and IPV stigma for male SCV victims. To retrieve male IT victims’ slope, IPV relationship type was re-coded (IT = 0, SCV = 1) and the regression model re-computed. Simple slopes for female subsamples of SCV and IT were calculated by repetition of steps.