Protective Factors associated with Resilience in Women Exposed to Intimate Partner Violence

Abstract

Objective

Research on adversity is often skewed toward assessing problematic functioning; yet many women display resilience following traumatic experiences. Examining individual, relational, community, and cultural variables can provide new knowledge about protective factors associated with resilience in women exposed to intimate partner violence (IPV). Controlling for demographics and circumstances of the violence, this study evaluated predictors of resilience, including spirituality, social support, community cohesion, and ethnic identity.

Method

The sample consisted of 112 women (M_age=32.12, SD=5.78) exposed to physical, psychological, and/or sexual intimate partner violence in the past 6 months. Approximately 70% of participants were Black. Hierarchical linear regression modeling was conducted to examine factors related to resilience. Model 1 included demographics (age, education, and socioeconomic status) and stressful life experiences. Model 2 added circumstances of the violence: IPV severity, IPV perpetration by participant, and number of violent partners. The third and final model added spirituality, social support, community cohesion, and ethnic identity.

Results

The final model was significant, F(11, 97)=6.63; p<.001, adj. R²=36.5%; with greater social support (β=.24; p=.009), more spirituality (β=.28; p=.002), and fewer violent relationships (β= −.25; p=.003) predicting higher resilience among women exposed to IPV.

Conclusion

While risk factors associated with IPV are well-researched, little is known about factors related to resilient functioning, especially among minority populations. Knowledge gained from this study can advance the field of violence research by its identification of potentially mutable variables related to resilience. Such research could be applied to developing strength-based interventions for at-risk populations of violence-exposed women.

Intimate partner violence (IPV), including physical violence, sexual violence, and psychological aggression by a current or former intimate partner, is a significant public health concern experienced by millions of women in the United States (Black et al., 2011). Over one in three women experience IPV in their lifetime with approximately seven million women experiencing IPV each year (Black et al., 2011). Racial/ethnic disparities in IPV victimization have been documented, with 54% of multiracial women, 44% of Black women, and 37% of Hispanic women endorsing lifetime IPV exposure, compared to 35% of White women (Black et al., 2011). Health consequences of IPV are substantial, with individuals exposed to violence endorsing depression, anxiety, suicidality, and worse overall health (Bonomi et al., 2006; Pico-Alfonso et al., 2006). To date, much of the research on IPV has been understandably skewed toward exploring the adverse effects of violence; however, a substantial proportion of women exhibit resilience following adversity (Jose & Novaco, 2015). Little empirical attention has been devoted to exploring resilience or the factors associated with higher resilience in IPV survivors. This study aims to utilize a social-ecological framework (Ungar, 2013) to examine individual, relational, community, and cultural factors related to resilience in IPV-exposed women.

Resilience has been conceptualized in a multitude of ways, from being viewed as the ability to bounce back and recover from stress (Smith et al., 2008), to positive growth following a traumatic experience (Tedeschi & Calhoun, 2004), and healthy functioning across personal and community contexts (Howell, Graham-Bermann, Czyz, & Lilly, 2010). These varying, and often conflictual, definitions have prompted the field to move toward a constructionist framework (Ungar, 2013), with recent theoretical conceptualizations highlighting the multidimensional and transactional processes across systems that underlie resilience (Howell & Miller-Graff, 2014).

One of the most significant reconceptualizations of resilience stems from Ungar’s proposed social-ecological model (Ungar, 2013). Under this framework, resilience is viewed as the capacity of individuals facing adverse circumstances to navigate their way to the psychological, social, cultural, and physical resources that sustain their well-being (Ungar, 2013). This definition highlights the impact of multiple systems on an individual’s capacity for resilience. Indeed, in their seminal work, Ungar and colleagues identified several interrelated factors associated with positive outcomes under stressful circumstances. These factors included relationships, a defined identity, power/control, social justice, access to material resources, a sense of cohesion/belonging, and cultural adherence; with the quality of these interrelated factors accounting for the majority of observed resilience among adolescents (Ungar et al., 2007).

Reflecting the impact of Ungar’s findings, many researchers have adopted social-ecological views of resilience, recognizing the importance of a comprehensive definition that incorporates relevant variables across multiple systems (Sanders & Munford, 2014; Theron, Liebenberg, & Malindi, 2013). In line with the growing movement to abandon oversimplified resilience models in favor of multilevel resilience (Anderson, Renner, & Danis, 2012), the current study is grounded in a social-ecological framework examining protective factors that may impact resilience among women exposed to recent IPV. Assessing resilience in this way acknowledges the variability in functioning following IPV exposure, reduces negative discourse surrounding IPV-exposed women, and highlights potentially modifiable factors that enhance resilience among a population in need of support.

While research on resilience among women exposed to IPV is quite limited, there is a clear need for a shift to examining more positive outcomes following adversity given consistent findings that many women exposed to IPV do not report enduring psychological problems (Anderson et al., 2012). Using a mixed methods approach, Anderson and colleagues (2012) reported that women who had previously experienced IPV emphasized the importance of secure relationships and informal and formal support. These women frequently endorsed taking pride in their achievements, relying on God for help, and believing that things happen for a reason. Women also reported personal, spiritual, and interpersonal growth throughout their journey following adversity (Anderson et al., 2012). Among women filing temporary restraining orders against male romantic partners in response to instances of IPV, Jose & Novaco (2015) observed that most women reported confidence in their personal strength, as well as their ability to: adapt to change, deal with whatever comes their way, bounce back from hardship, and achieve despite obstacles. While this qualitative information demonstrates that women endorse aspects of resilience in the midst of highly stressful experiences, it is crucial to understand what factors contribute to resilience in the aftermath of IPV. A multidimensional conceptualization of resilience may illuminate the complex, interrelated system of factors that bolster women’s resilience (Anderson et al., 2012).

Social-Ecological Protective Factors

Referencing prior work in related fields, we can identify potential protective factors within a social-ecological framework that may bolster resilience in women exposed to IPV. Influential protective factors can be conceptualized as individual characteristics (e.g., spirituality), relational qualities (e.g., social support), and community or cultural factors (e.g., community cohesion, ethnic identity). These variables are components of a complex, multilevel resilience model that may affect an individual’s ability to respond to stress and adversity.

Spirituality

Spirituality is often thought of as the belief in a higher power providing connection to life that transcends daily understanding (Decker, 1993). Modern definitions of spirituality expand this view to include feelings of harmony, peacefulness, and/or a relationship to someone or something (i.e. belief in the goodness of human nature or a spiritual guide) that provides forgiveness and contributes to psychological well-being (Koenig, 2010). Studies on spirituality in relation to mental health have been mixed (Seybold & Hill, 2001); however, recent research indicates that spiritual beliefs can be a source of comfort, hope, and meaning (Koenig, 2010). In the context of IPV, limited research suggests that spirituality may buffer against maladaptive health outcomes. For example, in a synthesis of qualitative findings across six studies, Yick (2008) reported that survivors of IPV drew strength from their spiritual beliefs in order to survive and cope with the violence. Survivors of IPV report using their individualized relationship with God as a lifeline or a relationship with dependability that they could not find elsewhere (Drumm et al., 2014). Additionally, women experiencing IPV have relied on their spirituality to find meaning, inner strength, wisdom, and compassion (Anderson et al., 2012). Of note, the majority of survivors of IPV derive strength from their personal, individual relationship with a spiritual connection rather than organized religious practices or institutional settings more commonly associated with the concept of religiosity (Drumm et al., 2014; Yick, 2008).

Spirituality has been framed as a culturally relevant protective factor among women of color (Raj, Silverman, Wingwood, & DiClemente, 1999). Specifically, Black women have reported using spirituality to cope with and find meaning in their adverse experiences, with higher levels of spirituality linked to more positive self-concept, coping, social support, attitudes towards parenting, and mental health (Mattis, 2002; Watlington, 2006). Further, Mitchell and colleagues (2006) found that Black women’s spiritual well-being mediated the link between IPV status and depressive symptoms, as well as the link between IPV status and parenting stress.

Social Support

Social support has consistently been shown to promote positive outcomes and reduce the likelihood of poor functioning in the midst of adversity (Agaibi & Wilson, 2005). Higher levels of social support have been linked to greater resilience amidst chronic daily stressors and potentially traumatic events (e.g., war, natural disaster, torture) (Agaibi & Wilson, 2005; Mhaka-Mutepfa et al., 2015). Conversely, lower levels of social support are associated with poorer mental health outcomes (Agaibi & Wilson, 2005). Unfortunately, compared to non-abused women, IPV-exposed women often report lower levels of perceived support from friends and family (Rodriguez et al., 2008). However, IPV-exposed women who maintain supportive ties are less likely to experience the adverse mental health consequences often endorsed in the aftermath of violence (Krauss, Wilson, Padrón, & Samuelson, 2016). Indeed, survivors of IPV have identified social support as a key resource that promotes positive functioning (Ahmad, Rai, Petrovic, Erickson, & Stewart, 2013). Among Black women, social support is an especially important culturally-relevant protective factor. Previous research has shown that Black women tend to rely on informal systems of support rather than formal systems (Raj et al., 1999), with higher levels of social support decreasing the negative impact of IPV on depression, anxiety, and parenting stress (Mitchell et al., 2006).

Community Cohesion

Community cohesion takes into account trust in neighbors, reciprocity, and community-wide connections to assess the level of social cohesion within a neighborhood or community (Greene, Paranjothy, & Palmer, 2015; Fagan, Wright, & Pinchevsky, 2014). Community cohesion buffers against poor mental health outcomes in the aftermath of trauma (Greene et al., 2015). Conversely, community disorder is related to higher rates of depression (Wright, Pinchevsky, Benson, & Radatz, 2015), greater posttraumatic stress (Butcher, Galanek, Kretschmar, & Flannery, 2015), and increased exposure to violence (VanderEnde, Yount, Dynes, & Sibley, 2012). Community cohesion has also been associated with resilience-building among adolescents exposed to violence (Fagan et al., 2014) and has served as a protective factor against psychological distress among adults exposed to frequent physical and psychological violence as children (Greenfield & Marks, 2010). Similarly, among Black individuals, community cohesion has been associated with lower levels of overall distress (Cutrona, Russell, Hessling, Brown & Murray, 2000), depressive symptoms (Nebbitt & Lombe, 2010), and PTSD symptoms (Gapen et al., 2011). Research on the relationship between community cohesion and resilience in IPV survivors is lacking.

Ethnic Identity

Ethnic identity is viewed as a dynamic construct defined by a positive sense of commitment, belonging, and involvement in one’s ethnic group (Avery, Tonidandel, Thomas, Johnson, & Mack, 2007). Developed throughout childhood and adolescence, ethnic identity has been associated with positive outcomes, such as increased self-esteem, coping, a sense of mastery, and optimism, as well as decreased loneliness and depression (Williams, Chapman, Wong, & Turkheimer, 2012). In contrast, an underdeveloped sense of ethnic identity, or a negative view of one’s ethnic identity, is associated with poorer psychological outcomes, such as low self-esteem and increased depression and anxiety (Williams et al., 2012). Thus, a positive, developed sense of ethnic identity can serve as a buffer against life stressors. Among Black individuals, the protective value of strong ethnic identity buffers against the negative effects of racial discrimination and other adversities (Li, Nussbaum, & Richards, 2007). A strong sense of ethnic identity has been linked to reduced suicidality (Kaslow et al., 2004) and lower depression (Lewin, Mitchell, Rasmussen, Sanders-Phillips, & Joseph, 2011) in women of color.

Violence-related Factors

While a comprehensive evaluation of protective factors across the social ecology offers novel information about resilience in IPV-exposed women, it is critical to also consider aspects of the violence itself that may impact functioning, including the number of violent relationships, violence severity, and women’s use of violent tactics toward her partner. Previous experiences of IPV significantly increase the likelihood of women entering a future violent relationship (Coker, Smith, McKeown, & King, 2000). Exposure to IPV from multiple partners escalates the risk for adverse mental and physical health consequences, with these women reporting high levels of shame, self-blame, depression, and substance use (Classen, Palesh, & Aggarwal, 2005). Prior research has also shown that greater violence severity may result in higher levels of depression, posttraumatic stress, and anxiety, as well as lower quality of life (Dutton et al., 2006; Straus et al., 2009). Bidirectional violence may occur in self-defense, defense of children, or retribution and control (Swan & Snow, 2006) and has been associated with overall poor functioning in women (Sprunger, Eckhardt, & Parrott, 2015). Further, Whitaker, Haileyesus, Swahn, & Saltzman (2007) indicate that women who engage in violent tactics against their partners have a significantly higher probability of injury and future violence. While extensive research has documented the deleterious effects of these violence-related risk factors, research on their relation to resilience is lacking in the literature.

Current Study

A growing body of research recognizes variability in functioning following adversity and acknowledges that many women exhibit resilience. Existing literature, however, has not considered protective factors across women’s social ecology that contribute to resilience following IPV, nor have studies concurrently assessed circumstances of the violence that may impact positive functioning. The limited literature on resilience in the context of IPV has not assessed the unique circumstances of women of color, despite research showing that Black women’s experience of violence differ from White women’s experience and the salience of resilience factors (e.g. spirituality, ethnic identity) vary across racial and ethnic groups. Thus, it is novel to simultaneously consider violence-related variables and factors at multiple social-ecological levels that contribute to resilience in a majority Black sample of women experiencing recent IPV. It is hypothesized that (1) aspects of the IPV (i.e., violence severity, women’s use of violent tactics toward her partner, and the number of violent relationships experienced by women) will be associated with lower levels of resilience, and (2) controlling for these circumstances of the violence; greater spirituality, social support, ethnic identity, and community cohesion will be positively related to resilience. By utilizing a social-ecological framework, we can begin to unpack the breadth of potentially modifiable factors that can enhance resilience.

The current study controls for the potential influence of demographic factors, including age, education, and socioeconomic status (SES), as well as other life stressors, on resilience. Specifically, studies have demonstrated positive associations between age and resilience (Cohen, Baziliansky, & Beny, 2014). Additionally, education may bolster resilience, with greater educational attainment associated with healthier functioning (Campbell-Sills, Forde, & Stein, 2009). Further, higher income levels tend to be associated with increased access to resources, more perceived control, and higher levels of resilience (Mhaka-Mutepfa et al., 2015), while lower SES is associated with an insecure sense of the future, passive coping, heightened stress, and poor health (McEwen & Gianaros, 2010). Finally, the cumulative number of lifetime stressful events one endures may influence individuals’ resilience, such that increased exposure to stressful events may undermine one’s resilience capacity (Alim et al., 2008).

Method

Participants

Participants included 112 women aged 22–49 (M_age = 32.12, SD = 5.78) who experienced IPV in the past six months. Women primarily identified as Black (67.8%) with 14.9% identifying as multi-ethnic, 11.6% as White, and 5.7% as other race/ethnicities. The majority of women (71.9%) endorsed an annual household income below $20,000; half were unemployed (50.4%).

Procedure

After obtaining IRB approval, women were recruited via flyers and direct staff referral from community organizations in the Midsouth U.S. serving individuals experiencing IPV. Participants were informed of the sensitive nature of questionnaires prior to study onset and were told they could skip any questions or discontinue participation at any time without penalty. After providing informed consent, women completed an hour-long semi-structured interview that was read aloud to them by trained research assistants. Participants were compensated for their time with a $20 gift card. All participants received a list of local and national mental health resources.

Measures

Demographics (Control variable)

Participants were administered a demographics questionnaire to obtain background information including: participant age, education, ethnicity, employment, household income, socioeconomic status, race, and relationship status.

Life Stressors (Control variable)

The Life Events Checklist (LEC) is a 17-item measure designed to screen for lifetime exposure to potentially traumatic events, including a natural disaster, transportation accident, other serious accident, assault, combat, other life-threatening event or serious injury, and sudden, unexpected death of a loved one (Gray, Litz, Hsu, & Lombardo, 2004). Participants reported whether each event happened to them and responses were summed to create a total cumulative life stressors score ranging from 0 to 17. In previous research, this measure has demonstrated adequate construct, convergent and discriminant validity as a stand-alone assessment of traumatic exposure (Gray, et al., 2004) among multiple ethnic groups (Lima et al., 2016). Reliability was not calculated for this measure because participants may experience one potentially traumatic event without necessarily experiencing another, so LEC items need not be related.

Number of Violent Partners (Risk predictor variable)

Women reported on their history of being in a violent relationship via a single item, “How many violent relationships have you ever been in?” Response options ranged from 0 to 4, with 0 representing no violent relationships and 4 representing more than 7 violent relationships.

IPV Perpetration (Risk predictor variable)

Women reported on their history of violence perpetration using 4-items adapted from the Revised Conflict Tactics Scale (CTS2; Straus et al., 1996). The History of Violence Perpetration Measure assessed the frequency with which participants were perpetrators of aggression toward their intimate partner over the preceding six months. Participants indicated the frequency of aggression (e.g., “Did you threaten or call your partner hurtful names?” and “Were you in a physical fight with your partner?”) on a scale from 0 (Never) to 3 (Many Times). Scores on the four items are summed to create a total score ranging from 0 to 12, with greater scores indicating more frequent violence perpetration. In the present study, convergent validity was established based on significant positive correlations with the CTS-2 subscale and total scores (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). The measure also showed acceptable internal consistency reliability of α = .70.

IPV Severity (Risk predictor variable)

Participants reported on the severity of the IPV they experienced over the past six-months using the Revised Conflict Tactics Scale (CTS2; Straus, Hamby, Boney-McCoy, & Sugarman, 1996; Straus, 1979). The CTS2 is a 78-item measure used to assess the severity of psychological, physical, and sexual violence experienced in a dating, cohabitating, or marital relationship. This study only included the 39 items that refer to violence perpetrated by the woman’s partner. Items inquire about the frequency of different types of IPV, including Physical Assault (e.g., “My partner beat me up.”), Psychological Aggression (e.g., “My partner swore or insulted me.”), Negotiation (e.g., “My partner showed care for me even though we disagreed.”), Injury (e.g., “I had a sprain, bruise, or small cut because of a fight with my partner.”), and Sexual Coercion (e.g., “My partner used threats to make me have oral or anal sex.”). Responses are scored on a 7-point scale, with 0 = never happened, 1 = happened once, 2 = happened twice, 3 = happened 3–5 times, 4 = happened 6–10 times, 5 = happened 11–20 times, and 6 = happened more than 20 times. Responses to the physical assault, psychological aggression, injury, and sexual coercion subscales are summed to create a total score in which higher scores reflect greater frequency of IPV. The CTS2 has good internal consistency reliability, with alpha coefficients ranging from .79 to .95, satisfactory construct and discriminant validity (Straus et al., 1996), and has been used with Black women (Fincher et al., 2015). In the present study, internal consistency was α = .95.

Spirituality (Protective predictor variable)

The Daily Spiritual Experiences Scale (DSES) is a 15-item self-report measure of personal interactions with God or a higher power across various dimensions of spirituality, such as personal intimacy with a higher power (e.g., “I feel God’s love for me directly”), strength and comfort (e.g., “I find comfort in my religion or spirituality”), perceived divine love (e.g., “I feel God’s love for me through others”), inspiration or discernment (e.g., “I am spiritually touched by the beauty of creation”), transcendence (e.g., “During worship, or at other times when connecting with God, I feel joy which lifts me out of my daily concerns”), and internal integration (Underwood, 2011; Underwood & Teresi, 2002). Items are assessed on a six-point Likert scale, ranging from never to many times a day. The DSES has high internal consistency reliability, with alpha coefficients of .94 to .95, adequate construct and discriminant validity (Underwood & Teresi, 2002), and has been used in studies with Black adults (e.g., Loustalot et al., 2011). In the present study, reliability was α = .94.

Social Support (Protective predictor variable)

The Lubben Social Network Scale Revised (LSNS-R) is a 12-item measure of perceived social support from family and friends (Lubben, Gironda, & Lee, 2002). It utilizes a six-point Likert scale ranging from 0 (least connected) to 5 (most connected), with items summed to create a total score where higher scores indicate more support. Sample items include, “How many of your [relatives/friends] do you see or hear from at least [once a month/once a week]?” and “How many [relatives/friends] do you feel at ease with, like you can talk about private or personal matters?” The LSNS-R has adequate internal consistency (α =.78) and adequate convergent validity with measures of mental and physical health (Lubben, Gironda, & Lee, 2002). It has also been used with minority racial/ethnic groups (Adams, Aranda, Kemp, & Takagi, 2002). In the present study, reliability was α = .89.

Community Cohesion (Protective predictor variable)

The Community Cohesion Scale (CCS) evaluates the cohesion and social ties within the participant’s communities. The 6-item measure was adapted from the Social Cohesion and Trust Scale developed by Sampson and colleagues (1997) with sample items, “In my community, people are willing to help their neighbors” and “People in my neighborhood can be trusted.” The CCS uses a four-point Likert scale from 1 (Strongly Disagree) to 4 (Strongly Agree). Scores range from 6 to 24 with higher scores reflecting stronger community cohesion. The original scale has been used with Black women and shows evidence of convergent validity with other measures of neighborhood SES and violence (Gapen et al., 2011). For the present study, reliability was strong α = .89.

Ethnic Identity (Protective predictor variable)

The Multigroup Ethnic Identity Measure-Revised (MEIM-R; Phinney & Ong, 2007) is a widely used measure of commitment and exploration of one’s ethnic identity. It includes 6 items that reflect the extent to which individuals seek out information about their ethnicity and are connected to their ethnic group. Sample items include “I have spent time trying to find out more about my ethnic group, such as its history, traditions, and customs” and “I understand pretty well what my ethnic group membership means to me.” Each item is rated on a five-point scale ranging from 1 (strongly disagree) to 5 (strongly agree) and scores are summed and averaged, with higher scores indicating greater identification with one’s ethnic group. The MEIM-R has good internal consistency, with alpha coefficients ranging from .83 to .89. Adequate construct and discriminant validity have been established (Phinney & Ong, 2007) and the measure has been used with various ethnic groups (Brown et al., 2014). In the current study, the MEIM-R demonstrated acceptable reliability (α = .87).

Resilience (Outcome variable)

The Connor-Davidson Resilience Scale (CD-RISC; Connor & Davidson, 2003) is a self-report measure consisting of 25 items that assess participant’s ability to respond to stress and adversity. Items represent major dimensions of resilience, including personal competence/high standards/tenacity (e.g., “You take pride in your achievements”), trust in one’s instincts (e.g., “In dealing with life’s problems, sometimes you have to act on a hunch without knowing why”), positive acceptance of change (e.g., “I am able to adapt when changes occur”), secure relationships (e.g., “I have at least one close and secure relationship that helps me when I am stressed”), control (e.g., You feel in control of your life), and influences from a higher power (e.g., “When there are no clear solutions to my problems, sometimes fate or God can help”). The measure utilizes a five-point Likert scale, ranging from not true at all to true nearly all of the time, with items summed to create a total score. The measure has established construct and discriminant validity and has been used with Black women (Brown, 2008). The overall scale internal consistency is α = .89, item-total correlations range from .30 to .70, and the interclass correlation coefficient is .87 (Connor & Davidson, 2003). In the present study, internal consistency was strong α = .91.

Data Analytic Plan

Analyses were completed in SPSS version 23. Hierarchical multiple regression analyses assessed relationships between each independent variable and the outcome of resilience. Model 1 included demographic control variables (i.e., age, education, SES), as well as other life stressors. Model 2 added circumstances of the violence, including IPV severity, IPV perpetration, and number of lifetime violent relationships. Model 3 added spirituality, social support, community cohesion, and ethnic identity to assess the influence of these protective factors on resilience, while controlling for demographics, other life stressors, and circumstances of the IPV. Multicollinearity diagnostics were examined using the variance inflation factor (VIF) and all values fell within an acceptable range (VIF < 2).

Results

All participants reported experiencing IPV in the past 6 months, with an average of 176.94 (SD = 139.68) violent events during that timeframe; thus women in this study endured roughly seven instances of IPV each week. Women primarily experienced psychological aggression, with 98.3% endorsing such incidents. Physical assaults were also frequently endorsed by women (91.5%). While sexual coercion (64.1%) and injuries (79.5%) were experienced by fewer participants. Violence perpetrated against women’s own partners (i.e., bidirectional violence) was reported by 85.7% of participants. Most women (57%) reported being in multiple violent relationships over their lifetime. See Table 1 for additional descriptive statistics, as well as correlations between continuous independent variables and resilience.

Table 1.

Means, Standard Deviations, and Correlations among Study Variables

	1	2	3	4	5	6	7	8	9	10	11	12
1. Resilience	-
2. Age	−.04	-
3. Education	.30**	.10	-
4. SES	−.12	.15	−.03	-
5. Life Stressors	.07	.09	.17	.15	-
6. Violent Partners	−.29**	.25**	−.03	.04	−.04	-
7. IPV Perpetration	−.10	−.05	−.15	−.01	−.06	.01	-
8. IPV Severity	−.12	−.06	−.13	.17	.29**	−.01	.30**	-
9. Spirituality	.44***	.05	.10	−.08	−.08	−.04	.08	−.23*	-
10. Social Support	.38***	−.05	.34***	−.23*	.08	−.13	.04	−.21*	.19*	-
11. Com. Cohesion	.16	.05	.33***	−.30**	−.02	−.03	−.09	−.10	.09	.31**	-
12. Ethnic Identity	.44***	−.02	.22*	−.20*	.02	−.14	−.05	−.03	.44***	.17	.22*	-
M	74.65	32.12	12.91	2.69	5.27	1.72	3.93	176.94	58.12	31.13	15.26	22.08
SD	13.91	5.78	2.30	0.98	2.72	0.77	2.90	139.68	14.34	12.59	5.12	5.55

N = 112.

^*p < .05.

^**p < .01.

^***p < .001

Model 1 of the hierarchical linear regression included demographic variables and a variable assessing exposure to other life stressors. This model was significant and accounted for 9.0% of the variance in resilience scores, F(4, 104) = 3.66; p = .008. In this model, only higher education was associated with more resilience (β = .30; p = .002), while age, SES, and total number of life stressors were not significant (See Table 2).

Table 2.

Summary of Hierarchical Regression Analysis Predicting Resilience

Variable	Resilience
	β	t	R2	ΔR2	F
Model 1			.090	--	3.66**
Age	−.108	−1.16
Education	.304	3.26**
Socioeconomic Status	−.121	−1.28
Stressful Life Events	.072	.75
Model 2			.189	.099	4.61***
Age	−.036	−.40
Education	.264	2.92**
Socioeconomic Status	−.083	−.91
Stressful Life Events	.073	.78
Number of Violent Relationships	−.332	−3.68***
IPV Perpetration	−.072	−.78
IPV Severity	−.113	−1.15
Model 3			.365	.176	6.63***
Age	−.032	−.39
Education	.144	1.61
Socioeconomic Status	.004	.04
Stressful Life Events	.049	.58
Number of Violent Relationships	−.252	−3.09**
IPV Perpetration	−.142	−1.70
IPV Severity	.001	.01
Spirituality	.283	3.13**
Social Support	.238	2.66**
Community Cohesion	−.032	−.36
Ethnic Identity	.173	1.90

^*p<.05,

^**p<.01,

^***p<.001

To examine the first study hypothesis, IPV severity, IPV perpetration, and lifetime number of violent partners were added to the demographics and life stressors model. This model was significant, F(7, 101) = 4.61; p<.001 and the amount of explained variance increased (Δ adj. R² = 9.9%). Lifetime number of violent partners was significantly related to current resilience, with having more violent partners associated with lower levels of resilience (β = −.33; p < .001). Higher education continued to be significantly linked to higher resilience (β = .26; p = .004).

To assess the second hypothesis, we added spirituality, social support, community cohesion, and ethnic identity to the third and final model. This third model was significant, F(11, 97) = 6.63; p<.001, and the amount of variance explained in resilience increased substantially (Δ adj. R² = 17.6%; adj. R² = 36.5%). Higher resilience was associated with more social support (β = .24; p = .009) and more spirituality (β = .28; p = .002). Notably, the lifetime number of violent relationships that women were in continued to be significantly associated with resilience (β = −.25; p = .003) in this final model.

Discussion

This study examined protective factors that may impact resilience in women experiencing IPV, including spirituality, social support, community cohesion, and ethnic identity, accounting for demographics, past life stressors, history of violence, IPV severity and IPV perpetration. While risk factors and psychopathology associated with IPV are well-researched, less is known about factors related to resilient functioning in the context of IPV. Further, our study population was unique involving primarily Black women in the Mid-South who all experienced recent (past 6 months) and severe (~7 IPV instances per week) intimate partner violence.

We hypothesized that IPV severity and perpetration, and number of violent partners, would be inversely related to resilience. However, only educational attainment and the number of violent relationships emerged as significant. Given literature on the positive role education plays in healthy functioning (Campbell-Sills, Forde, & Stein, 2009), it is not surprising that higher education attainment emerged as a predictor of higher resilience. Given research indicating that ethnic minority women have less access to higher education than their White counterparts (Sue & Sue, 2012), these findings are particularly salient for this population. It may be that cognitive resources gained via schooling serve to decrease stress, and thereby increase resilience. In addition, having less education may limit resources for coping with adversity, which may lead women to endorse having lower resilience. Given that causality cannot be inferred in these cross-sectional analyses, it may be that those women who are highly resilient are also more likely to display the perseverance needed to obtain higher levels of education.

The significant, inverse relationship between number of violent partners and resilience provides partial support for the first study hypothesis. The data suggests that the cumulative nature of IPV has a larger effect over other circumstances of the violence, including current IPV severity and bidirectional conflict. This finding aligns with the work of Foa et al. (2000) who found that IPV and impaired mental health interact in a vicious cycle whereby IPV leads to negative psychological outcomes that, in turn, place women at greater risk of revictimization by creating the belief that they are unable to limit future violence perpetrated against them. Women who are trapped in this cycle of violence may believe that they lack the ability to garner resources to improve their situation, which likely impacts their sense of resilience following multiple relationships characterized by violence.

In opposition to the first hypothesis, current IPV severity and perpetration were not significantly related to resilience. Given the high levels of current violence experienced by participants (seven IPV instances each week), it is likely that this uniformly high IPV exposure provided minimal variability to detect an effect on resilience. The rates recorded in this study reflect findings from Black and colleagues (2011) showing that women of color experience IPV at highly elevated levels. Similarly, over 85% of women endorsed bidirectional violence, likely limiting sample variability. Thus, future work should examine different frequencies of violence severity and perpetration, including more moderate levels of IPV. Notably, our measure of IPV perpetration did not gather details on the circumstances in which violence was used, such as in retaliation to being abused, such information is necessary to understand how victimization and perpetration interact to impact resilience. Despite these limitations, this study offers novel information to advance bidirectional IPV research.

The second study hypothesis assessed social-ecological protective factors related to resilience, while controlling for the effects of demographics, other life stressors, and circumstances of the violence. Although the potency of protective factors did not outweigh women’s history of violent relationships when predicting positive functioning, results suggest that spirituality and social support serve an important role in enhancing resilience. Thus, this study offers unique insight into how resilient functioning may be improved by both individual and relational protective factors, particularly for Black women. It may be that spirituality offers a sense of stability and comfort while enduring a violent relationship, which may lead to enhanced resilient functioning. Similar findings have been noted by Yick (2008), as well as Drumm and colleagues (2014) who found that spirituality bolstered one’s inner-strength and served as an additional source of support for survivors of IPV. Black women, in particular, have reported using religion and spirituality to cope with and find meaning in their experiences (Mattis, 2002). Therefore, spiritually-oriented strategies may be especially helpful in boosting resilience after IPV. Social support was also significantly related to resilience in this sample. As past research with Black women has shown, those with more social support display fewer negative outcomes, which further elucidates the need for women to maintain supportive ties both during and after adversity. Social support is likely providing an additional resource to IPV-exposed women, such as a warm and caring relationship that fosters effective coping strategies and improves psychological well-being to aid the amount of resilience displayed.

Surprisingly, community cohesion and ethnic identity were not found to be significant predictors of resilience in this study. It may be that women in violent relationships are isolated from their broader community networks and therefore not able to reap the rewards of a cohesive and supportive neighborhood. Unlike the proximal factors of spirituality and social support, these more distal factors of community cohesion and ethnic identity are less influential in resilient functioning. Consistent with previous literature and current results, Black women have a higher propensity to turn to more informal means of support, often family, friends, or their own spirituality rather than organized religion or formal counseling (Raj et al., 1999). It may be that women of color who are experiencing violence are also less likely to turn to their community due to fear of backlash or non-support, thus dampening their ability to form a deep connection to their community in this context. Another key explanation for this discrepancy may be related to power and control exhibited by violent partners. Specifically, if a violent partner limits access to transportation, resources, and time to connect with broad community networks and engage in personal exploration, then victimized women may not be able to develop a strong sense of ethnic identity and community.

Limitations

The cross sectional design precludes causal inference and the ability to investigate potential variability in resilient functioning over time as resilience may not be a stable construct. Given these concerns of causal inference, interpretations regarding the directionality of results cannot be confirmed. These findings are therefore exploratory, with future research urged to investigate causal pathways. The use of only self-report data introduces the potential for response bias, as participants may not have felt comfortable honestly responding to sensitive items. Additionally, the sample was entirely women, so no conclusions can be drawn about men experiencing IPV. Furthermore, the sample was recruited from community resource centers; therefore, these women may be different from individuals who have not sought formal support.

Clinical and Policy Implications

Results suggest that clinical approaches should be targeted toward those individuals who seem most isolated from central supportive figures (e.g., friends and family), as social support from was linked to more resilience. Culturally responsive interventions with IPV-exposed Black women should encourage reaching out to family and friends, as well as provide resources for expanding social networks. It is important for clinicians to convey respect for spirituality, regardless of the clinician’s own level of spirituality (Constantine, Lewis, Conner, & Sanchez, 2000). Clinicians should also consider spirituality under the modalities of “public” and “private,” which may be especially useful for Black women experiencing IPV. This examination should include a multicultural conceptualization of spirituality and how it may differ from religiosity or women’s willingness to seek out organized religion. Therapeutic interventions with highly spiritual women should help them foster beliefs of a deeper meaning of life, while also considering coping resources such as prayer and meditation (Gillum & Griffith, 2010). Results also highlight that IPV revictimization can dampen women’s ability to bounce back from hardship, more so than the severity of recent IPV or perpetration. Given these results, women with previous violent partners should be targeted for intervention. In addition to measuring IPV in the past six months, clinical interviews should incorporate ways of measuring lifetime IPV, perhaps by asking women to report information not only on their current violent partner but also on previous violent relationships.

Research Implications

Longitudinal examination of women exposed to IPV is needed to understand causal relationships among study variables and map out factors that can heighten resilience. Assessing multiple informants beyond the woman exposed to violence, such as children, parents, and others within women’s social circles, may provide important insight into the relationships that influence resilient functioning over time and who may also be impacted by the violence their loved one is experiencing. Additionally, while this study underscores the benefits of assessing cumulative stress and IPV exposure, there is much to be learned about past exposure to IPV, including influence of severity and type. Lastly, other protective factors such as self-esteem, optimism, and social competence should be investigated to determine their association with resilience in IPV-exposed women. Such studies may shed much-needed light on the protective factors at play when an individual has been exposed to violence and thus provide insights into strength-based intervention strategies.