Patterns of Help-seeking Strategies in Response to Intimate Partner Violence: A Latent Class Analysis

Abstract

Women engage in multiple strategies to cope with the impact of intimate partner violence (IPV). Prior research has focused predominantly on women’s service utilization and help seeking as individual acts, yet it is likely that women engage in distinct patterns of multiple help-seeking strategies to achieve safety. As such, the current article examines patterns of service-related help-seeking strategies survivors employ. This article also investigates demographic factors, relationship characteristics, and mental and physical health effects of IPV associated with patterns of help seeking. Using a web-based survey, data were collected from service-engaged adult female IPV survivors (n = 369) in the Southwest region of the United States. Latent class analysis (LCA), a person-centered analytical approach, was used to identify survivors’ patterns of help seeking. A 3-class LCA model was determined to be the best fit for the data. Among the sample, 50% of women broadly engaged formal and informal networks, 15% primarily engaged informal networks, and 35% broadly engaged networks but avoided legal systems while seeking other formal services. Findings indicated varying and significant associations between class membership and race/ethnicity, foreign-born status, number of children, IPV severity, and mental health symptoms. The findings reinforce the need for practitioners to be aware of the varied ways women choose or avoid seeking help and explore women’s preferences. Comprehensive and collaborative service networks are necessary for early detection and holistic care. Addressing structural factors is imperative for expanding the range of viable support options available to IPV survivors, particularly women of color.

Keywords

domestic violence intimate partner violence informal networks formal networks help seeking

Introduction

Women draw upon different resources to cope with intimate partner violence (IPV) and mitigate its harmful consequences (Coker et al., 2002; Dutton, 1992; Goodkind et al., 2004; Goodman et al., 2003; Shannon et al., 2006). Research suggests heterogeneity in women’s responses to IPV. Specifically, financial concerns, institutional challenges (e.g., police responding in unhelpful ways, distrust of law enforcement), stigma, and priorities inform women’s actions and help-seeking behaviors (Davies & Lyon, 2014; Hamby, 2014; Hamby & Gray-Little, 2007; Liang et al., 2005; Overstreet & Quinn, 2013). Using a person-centered analytic approach (Collins & Lanza, 2010), the current study adds to the existing literature by investigating patterns of women’s help-seeking strategies and analyzing women’s actions in response to IPV. Utilizing latent class analysis (LCA), this study also examines associations between help-seeking patterns and women’s demographic and relationship characteristics, as well as self-reported health and mental health symptoms.

Characteristics Associated with Help Seeking in Response to IPV

Significant research has focused on the conceptualization, measurement, and examination of the safety strategies women use to address IPV (Goodkind et al., 2004; Goodman et al., 2003). Help-seeking strategies are generally dichotomized and grouped as either informal and formal sources of help (Cho et al., 2020; Fanslow & Robinson, 2010; Leone et al., 2007; Meyer, 2010), although some research identifies additional categories, such as avoiding or placating the abusive partner as forms of active and passive resistance (Goodman et al., 2003). The majority of help-seeking research has focused on women’s utilization of specific formal services, such as protection orders (Durfee & Messing, 2012), counseling services (Henning & Klesges, 2002), and a combination of medical, legal, and social services (Anyikwa, 2015; Flicker et al., 2011; Lipsky et al., 2006; Macy et al., 2005; Nurius et al., 2011; Vatnar & Bjørkly, 2009).

Demographic Characteristics

Extant literature has identified a wide array of demographic, relationship, and health characteristics associated with survivors’ actions and help seeking in response to IPV. For example, prior research has shown mixed associations between race/ethnicity and help seeking. One study found higher rates of police intervention and obtaining a protection order among Black and Hispanic women compared to White women (Flicker et al., 2011), whereas another study found a lower likelihood of obtaining a protection order among Black women compared to White women (Durfee & Messing, 2012). Findings from Anyikwa (2015) revealed that one-third of Black women who experienced IPV did not engage the police as a help-seeking strategy within the past year; this was supported by women’s narratives of concern for their partners and anticipated mistreatment by law enforcement. In terms of non-legal resources, a number of studies found that Black women, relative to White women, are less likely to engage in mental health services or general formal help seeking (Cho et al., 2020; El-Khoury et al., 2004; Flicker et al., 2011; Henning & Klesges, 2002) and more readily utilize informal help seeking supports (Anyikwa, 2015; Cho et al., 2020). Additional research suggests Hispanic women, compared to White women, are less likely to seek housing assistance or health services (Flicker et al., 2011; Lipsky et al., 2006).

With respect to socioeconomic status, findings again are mixed. Several studies found associations between household income and education with increased use of mental health and legal services (Durfee & Messing, 2012; Flicker et al., 2011; Henning & Klesges, 2002). Other studies, however, did not find a significant association between help seeking and socioeconomic status (Duterte et al., 2008; Hanson et al., 2019). It is not clear whether these discrepancies emerged because of the sample, analytic approach, or other factors.

As to other demographic factors, being foreign-born was found to be associated with decreased formal help seeking (Cho et al., 2020). Being older appears to be associated with increased use of mental health and legal services (Durfee & Messing, 2012; Flicker et al., 2011; Henning & Klesges, 2002), and general formal help seeking (Cho et al., 2020), but not seeking help informally from family members (Flicker et al., 2011; Leone et al., 2007). Finally, having a child shapes survivors’ actions in response to IPV. Prior research shows that mothers consider the effect of IPV on their children and, in turn, are more likely to engage in formal help-seeking processes (Randell et al., 2012). Having a child is associated with increased use of mental health and legal services (Durfee & Messing, 2012; Duterte et al., 2008; Henning & Klesges, 2002), as well as high overall engagement in safety strategies (Hanson et al., 2019).

Relationship Characteristics

IPV severity is associated with increased formal help seeking, including increased use of mental health, medical, and legal services (Cho et al., 2020; Duterte et al., 2008; Flicker et al., 2011; Henning & Klesges, 2002; Macy et al., 2005). Survivors who experienced intimate terrorism (i.e., IPV embedded in a general pattern of control; Johnson & Leone, 2005) were less likely to seek informal support and more likely to seek help from the police, medical organizations, and counselors when compared to women categorized as victims of situational couple violence (Leone et al., 2007).

There are conflicting findings, however, with regard to relationship status, with women whose partner is a current/former spouse (vs. a dating partner) seeking help from police less often, but with no differences or higher rates of other forms of formal and informal help seeking (Flicker et al., 2011; Henning & Klesges, 2002; Leone et al., 2007). While one study found that women in longer relationships were less likely to engage in formal help seeking and more likely to engage in informal help seeking (Leone et al., 2007), another found no differences by relationship duration (Duterte et al., 2008).

Mental and Physical Health

Research has consistently shown that negative mental health symptoms (e.g., post-traumatic stress disorder (PTSD), depression) are associated with increased formal help seeking (Cho et al., 2020; Hanson et al., 2019; Leone et al., 2007). However, studies have not found consistent associations between help seeking and women’s physical health. Specifically, one study found an association between physical health functioning and decreased formal help seeking (Macy et al., 2005) while another did not identify differences by physical health status (Cho et al., 2020).

Help-seeking Patterns

Although empirical studies have investigated help seeking in response to IPV, important discrepancies in evidence persist when examining particular types of services or individual forms of help seeking. Taken together, these studies identify that distinct behavioral, characteristic, and relationship history factors are associated with service and support options. This suggests distinct patterns of relationships among factors embedded within populations (Nurius et al., 2011) and a need to further investigate survivor patterns of help seeking. The diverse findings also suggest heterogeneity in women’s responses to IPV and elucidate the challenges inherent in capturing women’s complex responses to IPV with variable-oriented analytic approaches (Bergman & Trost, 2006).

Person-oriented methods provide a distinct methodological advantage because survivors are placed in mutually exclusive groups based on similar patterns of behavior, providing useful comparative information about how survivors make choices when seeking help and engaging systems. For example, Ben-Porat (2020) conducted latent class analysis with a sample of 499 women residing in shelters in Israel to identify patterns of service utilization. Results of the study revealed a 3-class model: Substantial Use (n = 46, 9%), Welfare and Criminal System (n = 278, 56%), and Minimal Use (n = 175, 35%). It was also found that the women who identified as Israeli-Arab (vs. Israeli-Jewish), experienced more frequent IPV and increased PTSD symptoms were more likely to be in the Substantial Use class (vs. Minimal Use). Conducting cluster analysis with non-help-seeking population in the United States (N = 725, recruited through advertisements), Hanson et al. (2019) identified three clusters of safety strategies: Exploring Safety Options (n = 190, 27%, characterized by the fewest safety behaviors), Avoiding the Justice System (n = 251, 35%), and Trying Everything (n = 273, 38%). Results of this study revealed significant differences in women’s characteristics by clusters. Specifically, women having a child below age 18, experiencing more IPV and PTSD symptoms were more likely to be in the Trying Everything class (vs. Exploring Safety Options or Avoiding the Justice System). These studies mark an important step forward and point to the need for person-oriented research with diverse samples to understand of survivors’ responses to IPV.

Conceptual Framework

The current analysis draws from theoretical frameworks that seek to elucidate women’s actions and help seeking in response to IPV (Gondolf & Fisher, 1988; Goodman et al., 2003; Hamby, 2014; Liang et al., 2005). Contrary to earlier theories that depict survivors as passive in the face of abuse (e.g., learned helplessness; refer to Walker, 1979), contemporary studies demonstrate that women, faced with intensifying violence, expand their strategic efforts in response to IPV (Goodman et al., 2003). The Survivor Hypothesis proposed by Gondolf and Fisher (1988), for example, uses empirical research to contradict the learned helplessness frame and suggests that IPV survivors usually develop a wide array of protective strategies to cope with IPV. Similarly, Hamby and Gray-Little (1997, 2007) proposed the Competency Model—in contrast to the Deficit Model that emphasizes IPV survivor’s cognitive and behavior deficiencies including minimizing violence—arguing that survivors usually report more active (e.g., calling the police) instead of passive (e.g., being silent) behavior responses as severity of IPV increases.

Liang et al. (2005) illustrate women’s help-seeking process from problem recognition and definition, decision to seek help, to support selection and articulate how individual, interpersonal, and sociocultural factors may shape actions survivors take to seek help. For example, survivors’ perceptions of police responses as ineffective, stigmatizing, inadequate, culturally inappropriate, or harmful may decrease their likelihood of seeking help from the police (Decker et al., 2020; Liang et al., 2005; Overstreet & Quinn, 2013). Social status, language skills, access to financial resources, and social support can serve to enable survivors’ help seeking; the lack thereof, such as for new immigrants, can severely impede options available to women in seeking respite from violence (Liang et al., 2005). Geography, such as being in a rural location or far from services, may also influence the options available to survivors. The intersection of these factors with social positions associated with race/ethnicity, gender identity, and sexual orientation (Crenshaw, 2017) adds further complexity and nuance to understanding the spectrum of actions survivors take. IPV survivors with marginalized social positions face structural racism and discrimination that further shape daily survival strategies.

Collectively, contemporary theories that highlight women’s use of protective strategies point out that women take actions to stop, prevent, or escape IPV in multiple nuanced ways, seeking help from formal service systems, informal systems, and engaging in less visible coping strategies to protect themselves, children, family, and pets from an abusive partner (Hamby, 2014). With the aim of exploring distinct patterns of help-seeking strategies, this study approaches help seeking from an empowerment standpoint that recognizes the difficulties inherent in seeking safety and acknowledges the heterogeneity among survivors and their responses to IPV shaped by a multitude of factors and forces at play.

Research Questions

Previous person-oriented examinations of help-seeking have examined patterns of help-seeking among service-engaged women outside the United States or have recruited survivors from the broader community. In light of the heterogeneity in women’s responses to IPV, the current exploratory study aims to add to the existing literature by using a person-centered approach to understand patterns of help-seeking behavior and predictors of those patterns with a sample of survivors engaged with IPV services at the time of data collection. This research seeks to expand the understanding of patterns of help seeking among U.S. women likely to present to IPV services in order to inform IPV service delivery and future research.

The current article addresses three research questions: (a) To what extent do service-engaged survivors employ strategies and actions to address violence in their intimate relationships? (b) What are patterns of help-seeking strategies and actions? (c) What demographics, relationship characteristics, and health factors are associated with identified patterns of help seeking with a service-engaged sample of survivors? To our knowledge, extant literature has not examined help-seeking patterns among a sample of service-engaged IPV survivors in the United States. Therefore, the current study does not have prior hypotheses on the number of patterns that would be identified.

Methods

The current analysis draws on baseline data (N = 377) collected from a multiyear study conducted in 2016–2018 examining the effectiveness of myPlan (www.myplanapp.org), an online safety decision aid (Eden et al., 2015; Glass et al., 2015), among survivors seeking domestic violence services at the time of data collection. Eight participants had missing responses on all variables that measured help seeking and were excluded from the analytic sample (n = 369).

Recruitment

Recruitment targeted women currently engaged with IPV/domestic violence services. Participant recruitment took place between April 2016 and August 2018 at 16 domestic violence service agencies such as shelters and agencies providing supportive services and case management in one Southwest state of the U.S. Researchers offered the myPlan intervention and study as a part of intake procedures, in common areas, and during client group meetings. To be eligible for the study, women had to: (a) be 18 years or older, (b) be comfortable using a computer or tablet, (c) be English speaking/literate, (d) have reported intimate partner abuse in the past 6 months (e.g., physical abuse or threats, forced or coerced sexual abuse, made to feel unsafe by an intimate or ex-intimate partner), and (e) have safe access to the internet and an email account (defined as an account used exclusively by the survivor with no history of sharing the password). If an individual did not have a safe and active email account and wanted to participate, a research assistant helped the participant create a new account.

Data Collection Procedures

For women who expressed interest in the study, researchers provided access to an iPad or desktop computer to review study related content. Research assistants were available to answer questions, assist participants with navigating the online platform, read the survey aloud upon request, and provide online safety planning information. Women who were not eligible or declined to participate in the study were given the option for accessing the myPlan intervention only. The survey took an average of 60-minutes to complete and women received a $20 gift card for participating.

Measures

Background.

Participants self-reported their age, race/ethnicity, U.S. nativity, employment status, education level, household income, number of children under age 18, whether they lived with the abusive partner before seeking services, and relationship duration.

Strategies and actions in response to IPV.

The current article uses 11 items from the Intimate Partner Violence Strategies Index (Goodman et al., 2003), organized into 7 categories of safety strategies: (a) engaging informal networks, (b) seeking help for their abusive partner (e.g., substance abuse treatment, counseling), (c) staying in a shelter, (d) seeking legal services, (e) engaging police, (f) seeking housing assistance to move to a safe location, and (g) visiting a health care provider. Some items were combined as they measure similar safety strategies. For example, items that asked women whether they had (a) stayed with family or friends to escape from IPV, (b) talked with family, friends, neighbors, or a co-worker/boss about IPV, and (c) sent kids to stay with a friend or family member were combined into a variable named engaging informal networks. All variables were categorized as dichotomous (yes/no) to indicate whether the participant used that safety strategy.

Self-reported health and mental health.

Physical health was measured by a question that asked the participants to self-rate their health status. Responses were dichotomized as good/very good/excellent health or fair/poor health.

PTSD and depression were measured using standardized instruments. The PTSD CheckList-Civilian Version (PCL-C) measures symptomology of PTSD for the civilian population as well as generalized traumatic events (Conybeare et al., 2012; US Department of Veteran Affairs, 2012). The instrument comprises 17 items, is rated on a 5-point Likert scale, and measures the degree to which the individual experiences symptoms within the past month (i.e., not at all to extremely). This study used the composite score of the PCL-C with each item response being aggregated to create a total severity score (ranging from 17–85; higher scores representing greater reported post-traumatic stress symptoms). This widely utilized instrument demonstrates strong internal consistency in the analytic sample (α = .93).

The Center for Epidemiologic Studies Depression Scale Revised (CESD-R) measures depressive symptoms based on those outlined in the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-V; Eaton et al., 2004). The measure is comprised of 20 items rated on a 4-point Likert scale (i.e., not at all or less than one day to nearly every day for two weeks). This study used the composite score of the CESD-R with each item response being aggregated to create a composite score (ranging from 0–60; higher scores representing greater reported depressive symptoms). This measure demonstrates strong internal consistency in the analytic sample (α = .94).

Intimate partner violence.

The Composite Abuse Scale (CAS) was used to measure severe abuse from an intimate partner. CAS is a 30-item instrument and is rated on a 5-point Likert scale (i.e., never to daily). This instrument comprises four subscales that measure different types of intimate partner abuse occurring in the last 6 months, including physical abuse (7 items), harassment (4 items), emotional abuse (11 items), and severe abuse factors (8 items; Hegarty & Valpied, 2013). This study utilized the Severe Combined Abuse Subscale (scores ranging from 0–40). The severity scale has good internal consistency for the analytic sample (α = .84). The Women’s Experiences with Battering Scale (WEB) is a 10-item instrument measuring women’s perceptions of the ways in which abuse has impacted her life (e.g., feeling controlled) (Smith et al., 1995). Scores range from 10 to 60 and the scale was high for internal consistency for the analytic sample (α = .91).

Data Analysis

LCA was used to examine patterns of strategies and actions that this sample of women has taken in response to IPV. LCA is a person-centered analytic approach that uses observed categorical indicator variables to identify unobserved homogeneous subgroups comprising similar individuals or cases (Collins & Lanza, 2010; Wang & Wang, 2012). By applying an LCA approach, women who report similar behaviors or patterns in terms of their help-seeking strategies and actions are placed into distinct “classes” or groups based on estimated posterior membership probabilities. To guide model selection (appropriate number of classes) and assess model fit, multiple statistics and statistical tests were performed including the Akaike’s information criterion (AIC), Bayesian information criterion (BIC), Lo–Mendell–Rubin likelihood ratio (LMR LR) test, adjusted Lo–Mendell–Rubin likelihood ratio (ALMR LR) test, and bootstrap likelihood ratio test (BLRT; Wang & Wang, 2012). In applying these tests, smaller values indicate better model fit and a significant p-value (p < .05) of the LMR LR or BLRT test indicates that a model with one additional class is a better fit than a model with one less class (Wang & Wang, 2012). LCA was conducted with Mplus 8 (Muthén & Muthén, 1998–2017).

In order to conduct LCA, seven binary indicator variables were first created to capture women’s help-seeking actions and strategies. The variable value of 1 was assigned if the participant responded “yes” to engaging in the strategy/action, whereas, the variable value of 0 was assigned if the participant responded “no” to having engaged in that specific strategy/action. LCA procedures were then applied in order to identify patterns of women’s help-seeking behaviors. To further investigate the association between latent class membership and demographics, relationship characteristics, and health-related characteristics, a three-step latent class modeling was adopted, a widely used analytic approach in LCA analysis (Asparouhov & Muthén, 2018; Bakk et al., 2013; Bolck et al., 2004; Vermunt, 2017). More specifically, we conducted equality tests of means across classes by using the auxiliary BCH command in the Mplus software package to investigate the association between latent class membership and the individual and interpersonal factors specified (Asparouhov & Muthén, 2018). The BCH approach, which uses weights to reflect the measurement error of the latent class variable, is one of the bias-adjusted stepwise approaches to latent class modeling and is more robust than other estimating approaches (refer to Asparouhov & Muthén, 2018; Bakk & Vermunt, 2016). The current study used the automatic version in programming (as opposed to the manual version), allowing evaluation of the mean of a distal variable across classes (Asparouhov & Muthén, 2018). Regarding missing values, Mplus accounts for missing data on any latent class indicator variables (i.e., the 7 binary variables) using full information maximum likelihood estimation (Muthén & Muthén, 1998–2017). Mplus does not account for missing data for observations that had missing values for all indicator variables, thus excluding 8 participants (n = 369) in the present analysis.

Table 1 presents fit statistics for latent Classes 1–5. As shown, the lowest BIC was for the two-class model, and the lowest adjusted BIC (ABIC; 3085) was for the three-class model, suggesting support for either a two or three-class model. However, results of the LMR, ALMR test, and BLRT suggested that the three-class model performed better than the two-class model. Although AIC also lends some support for a four-class model, which had the lowest AIC—the three-class model was determined to be the best-fit model as the BIC is typically used when deciding on the number of classes (Nylund et al., 2007). Finally, entropy, which summarizes certainty in class separation and posterior classification, was .85 for the 3-class solution, also suggesting that it is a superior LCA model (Ramaswamy et al., 1993).

Table 1.

Comparisons of Different LCA Models.

Model	AIC	BIC	ABIC	Entropy	LMR LR TestP-value	ALMR LR TestP-value	BLRTP-value	Absolute Frequency for Smallest Class	Relative Frequency for Smallest Class
1-class LCA	3217	3244	3222	–	–	–	–	–	–
2-class LCA	3095	3154	3106	.589	<.001	<.001	<.001	174.32	.47
3-class LCA	3068	3158	3085	.849	<.005	<.005	<.001	56.10	.15
4-class LCA	3062	3184	3085	.824	NS	NS	<.05	59.63	.16
5-class LCA	3065	3217	3093	.824	NS	NS	NS	8.04	.02

Note. –: Not applicable; NS = Non-significant; AIC = Akaike’s information criterion; BIC = Bayesian information criterion; ABIC = Adjusted BIC; LMR LR test = Lo–Mendell–Rubin likelihood ratio test; ALMR LR test = Adjusted Lo–Mendell–Rubin likelihood ratio test; BLRT = Bootstrap likelihood ratio test.

Results

Descriptive Analysis

Table 2 presents the sample characteristics. Participants were between 18 and 64 years of age (M = 38, SD = 11). The majority were non-White (61%), U.S.-born (88%), and unemployed (69%). Slightly more than half of the sample completed some college or more (n = 201, 54%). Most women lived with the abusive partner before seeking services (76%) and had endured the abusive relationship for more than two years (66%). Out of the 369 participants, more than half (59%) had at least one child under age 18, with a median of 2 children among the 59% of women who had at least one child under age 18.

Table 2.

Sample Characteristics.

Categorical Variable	%	n	N (Missing %)
Race/ethnicity			369 (0)
White	39	145
Black	24	87
Hispanic	12	46
Other (American Indian, Asian, multi-racial)	25	91
U.S. nativity			357 (3)
U.S.-born	88	324
Foreign-born	9	33
Employment status			353 (4)
Employed	27	98
Not employed	69	255
Education level			357 (3)
High school diploma and below	42	156
College and above	54	201
Monthly household income			347 (6)
$500 and below	56	208
$501 and above	38	139
Self-rated health			345 (7)
Good, very good, or excellent	62	227
Fair or poor	32	118
Lived with abusive partner before shelter			355 (4)
Yes	76	281
No	20	74
Abusive relationship duration			356 (4)
Less than 2 years	30	111
Over 2 years	66	245
Continuous Variable	Mean	SD	N (missing %)
Age (in years)	37.71	10.99	368 (0)
Number of children under age 18	1.50	1.82	369 (0)
PCL composite score	57.56	17.60	322 (13)
CESD-R composite score	38.16	15.97	322 (13)
CAS Severe Abuse Subscale score	10.81	8.89	332 (10)
WEB composite score	51.76	10.39	335 (9)

Note. PCL-C = The PTSD CheckList-Civilian Version; CESD-R = The Center for Epidemiologic Studies Depression Scale Revised; CAS = The Composite Abuse Scale; WEB = The Women’s Experiences with Battering Scale.

With regard to strategies used by this sample of women in response to IPV, more than half of participants had engaged their informal networks for help (87%, n = 320), sought help for their abusive partners (e.g., substance abuse treatment, counseling, 66%, n = 238), stayed in a shelter (53%, n =194), or sought legal services (e.g., criminal charges, legal aid, 52%, n = 191). Nearly half of the participants had sought assistance to move to a safe location (49%, n = 181). Fewer women had engaged police (43%, n = 160) or visited a health care provider (40%, n = 149).

Latent Class Analyses

Table 3 and Figure 1 present the three-class LCA model based on model estimated probabilities of endorsing the seven indicators of help seeking. Within this analysis, women were placed in three distinct classes (mutually exclusive groups) based on patterns of help seeking as follows: (a) Broadly Engages Formal and Informal Networks; (b) Primarily Engages Informal Networks; and (c) Broadly Engages Networks but Avoids Legal Systems.

Figure 1.

Three-class model.

Table 3.

Item-response Probabilities.

Variable	Latent Class
	Class 1 Broadly Engages Formal and Informal Networks(n = 183^a)	Class 2 Primarily Engages Informal Networks(n = 56)	Class 3 Broadly Engages Networks but Avoids Legal Systems(n = 130)
Latent class prevalences (%)	50	15	35
Item-response probabilities corresponding to a response of positive endorsement
Previously stayed in a shelter	.63	.00	.62
Sought housing assistance	.58	.09	.55
Visited health care provider	.59	.01	.35
Engaged the police	.67	.20	.25
Sought legal services	1.00	.29	.00
Sought help for their abusive partners	.79	.24	.67
Engaged informal networks for help	.96	.70	.86

Note. ^aClass counts and proportions for latent classes are based on estimated posterior probabilities. Class counts based on probabilities were round to nearest integer to reflect original sample size.

^bItem-response probabilities >.5 in bold to facilitate interpretation.

Class 1, Broadly Engages Formal and Informal Networks, comprised 50% (n = 183) of the sample. This class was characterized by high probabilities of endorsing all indicators with over 50% engagement in each strategy: informal networks (96%), sought help for their abusive partners (79%), previously stayed in a shelter (63%), sought legal services (100%), sought housing assistance (58%), engaged the police (67%), and visited a health care provider (59%). Across each of these strategies, women in this class had a greater probability of utilizing each type of help-seeking compared to the average probability of the sample. For example, every woman in the Broadly Engages Formal and Informal Networks class had sought legal services compared to 52% in the sample. Similarly, 96% of them had engaged informal networks for help compared to 87% in the sample. Figure 1 provides a visual illustration of the probabilities of engaging in each strategy by class and compared to the sample mean.

Class 2, Primarily Engages Informal Networks, comprised 15% (n = 56) of the sample. This class was characterized by low probabilities of endorsing every type of help-seeking compared to the sample average: informal networks (70%), sought help for their abusive partners (24%), sought legal services (29%), sought housing assistance (9%), engaged police (20%), and visited a health care provider (1%). Most notably, no survivors in this class had previously stayed in a shelter (0% probability). Although this group primarily sought help through their informal support networks, they utilized their informal networks less than average (70% vs. a sample average of 87%), demonstrating lower help-seeking in general.

Class 3, Broadly Engages Networks but Avoids Legal Systems,¹

We named Class 3 as Broadly Engages Networks but Avoids Legal Systems in recognition that IPV survivors may never experience justice from criminal justice systems. Legal systems here refer to legal services as well as the police.

comprised 35% (n = 130) of the sample. This class was characterized by average and above average probabilities of endorsing most help-seeking strategies: informal networks (86%), sought help for their abusive partners (67%), had previously stayed in a shelter (62%), and sought housing assistance (55%). Figure 1 demonstrates that the pattern of using these social services was parallel with those of the Broadly Engages Formal and Informal Networks class. In contrast, however, this group had lower than average probabilities of calling the police (25%) and seeking legal services (0%). Visiting a health care provider (35%) was also used comparatively less often than Class 1 but more often than Class 2, reflecting a probability closer to the sample mean (40%).

Bivariate Associations with Class Membership

Table 4 summarizes results of the bivariate analyses on associations between latent class membership and demographic, relationship-, and health-related variables.

Table 4.

Comparison Between Latent Classes: Results of Bivariate Analysis.

Proportion/Mean (SE)	Class 1	Class 2	Class 3	Overall Test
Demographic Characteristics
Race/ethnicity
White	39% (4%)	39% (9%)	40% (5%)	χ²(2) = .01, NS
Black	20%^a (3%)	21% (7%)	31%^b (5%)	χ²(2) = 4.16, NS
Hispanic	14% (3%)	16% (6%)	9% (3%)	χ²(2) = 2.31, NS
Other (American Indian, Asian, multi-racial)	27% (3%)	24% (8%)	21% (4%)	χ²(2) = 1.42, NS
Foreign-born	13%^a (3%)	13% (6%)	3%^b (2%)	χ²(2) = 9.28, p < .05
Employed outside of home	25% (3%)	37% (9%)	27% (4%)	χ²(2) = 1.57, NS
College and above	57% (4%)	56% (9%)	55% (5%)	χ²(2) = .16, NS
Monthly household income as $501 and above	40% (4%)	35% (10%)	42% (5%)	χ²(2) = .45, NS
Age (in Years)	37.25 (.77)	38.70 (2.27)	37.94 (1.13)	χ²(2) = .56, NS
Number of children under age 18	1.76^a (.15)	.84^b (.23)	1.43 (.18)	χ²(2) = 11.27, p < .005
Relationship Characteristics and IPV Experiences
Lived with abusive partner before shelter	76% (3%)	79% (7%)	83% (4%)	χ²(2) = 1.74, NS
Abusive relationship duration over 2 years	70% (4%)	57% (9%)	72% (5%)	χ²(2) = 2.09, NS
CAS Severe Abuse Subscale	12.90^a (.77)	6.27^b (1.33)	9.84^c (.81)	χ²(2) = 20.61, p < .001
WEB composite score	54.37^a (.71)	44.67^b (2.34)	51.21^c (1.11)	χ²(2) = 20.16, p < .001
Self-reported Health and Mental Health
Self-rated health as good and above	66% (4%)	62% (9%)	67% (5%)	χ²(2) = .21, NS
PCL composite score	61.40^a (1.39)	50.27^b (3.85)	54.97^b (1.74)	χ²(2) = 13.91, p < .005
CESD-R composite score	41.37^a (1.27)	31.11^b (3.17)	36.37^b (1.63)	χ²(2) = 12.96, p < .005

Note. Class 1 = Broadly engages formal and informal networks; Class 2 = Primarily engages informal networks; Class 3 = Broadly engages networks but avoids legal systems; PCL-C = The PTSD CheckList-Civilian Version; CESD-R = The Center for Epidemiologic Studies Depression Scale Revised; CAS = The Composite Abuse Scale; WEB = The Women’s Experiences with Battering Scale; NS = Non-significant.

Subscripts indicate significance differences in pairwise comparisons at p < .05.

Demographic characteristics.

The percentage of Black respondents (31% vs. 20%) was higher in the Broadly Engages Networks but Avoids Legal Systems class than the percentage in the Broadly Engages Formal and Informal Networks class, whereas the percentage of foreign-born individuals (13% vs. 3%) was higher in the Broadly Engages Formal and Informal Networks class than in the Broadly Engages Networks but Avoids Legal Systems class. Compared to women in the Broadly Engages Formal and Informal Networks class, women in the Primarily Engages Informal Networks class had fewer children under age 18 (.84 vs. 1.76). There were no differences between classes on all other demographic variables (i.e., employment status, education level, age, income).

Experiences of intimate partner violence.

Women in the Broadly Engages Formal and Informal Networks class reported more frequent IPV, including more negative effects of IPV, compared to women in either the Primarily Engages Informal Networks class or the Broadly Engages Networks but Avoids Legal Systems class (CAS score = 12.9 vs. 6.27 or 9.84; WEB score = 54.37 vs. 44.67 or 51.21). In addition, women in the Broadly Engages Networks but Avoids Legal Systems class experienced more frequent IPV, and reported more negative effects of IPV, compared to those in the Primarily Engages Informal Networks class (CAS score = 9.84 vs. 6.27; WEB score = 51.21 vs. 44.67). There were no differences between classes regarding other relationship characteristics with the abusive partner (i.e., lived with abusive partner, relationship duration).

Self-reported health and mental health.

Women in the Broadly Engages Formal and Informal Networks class reported higher levels of PTSD and depressive symptoms than women in either the Primarily Engages Informal Networks class or the Broadly Engages Networks but Avoids Legal Systems class (PCL-C score = 61.4 vs. 50.27 or 54.97; CESD-R score = 41.37 vs. 31.11 or 36.37). Women in the Primarily Engages Informal Networks class had similar levels of PTSD and depression as women in the Broadly Engages Networks but Avoids Legal Systems class. There were no significant differences between classes with regard to self-reported physical health.

Discussion

This article examined patterns of help-seeking strategies tried by survivors engaged with IPV services at the time of data collection. The findings revealed the extent to which this sample of service-engaged survivors actively sought help and employed strategies to enhance their safety and wellbeing. The analysis identified three broad classes of help seeking: (a) Broadly Engages Formal and Informal Networks (50%), (b) Primarily Engages Informal Networks (15%), and (c) Broadly Engages Networks but Avoids Legal Systems (35%). Importantly, the findings revealed associations between class membership and race/ethnicity, foreign-born status, number of children, IPV severity, and mental health symptoms. The findings contribute to a nascent body of research that examines patterns of help seeking among IPV survivors (Ben-Porat, 2020; Hanson et al., 2019), further reiterate degrees of heterogeneity in women’s responses to IPV (Davies & Lyon, 2014; Hamby, 2014), and offer new insights for research and practice.

Of the two previous studies examining patterns of women’s help seeking (Ben-Porat, 2020; Hanson et al., 2019), each identified three classes and found a group of women who used few formal services, consistent with but larger than the Primarily Engages Informal Networks class found in this study. Similarly, both previous studies also identified a high service utilization group like the Broadly Engages Formal and Informal Networks group in this study, characterized by high levels of endorsing all help-seeking strategies. Although the categories were similar, we found a higher proportion of participants (50%) in the high service utilization group compared to 9% in the study conducted in Israel (Ben-Porat, 2020) and 38% in another U.S. sample (Hanson et al., 2019). The two U.S.-based studies diverged the most with the Israeli sample with regard to the final pattern of help seeking. Similar to the current study, the other U.S.-based sample of women avoided the legal system as a help-seeking strategy (Hanson et al., 2019). In the study conducted in Israel (Ben-Porat, 2020), women’s third pattern involved engaging the legal system in combination with the social service system. Together, these findings suggest a pattern of service seeking among women in the United States shaped, in part, by concerns about the legal system (e.g., discordance between women’s justice preferences and expectations and the current practice of legal systems; refer to Decker et al., 2020; Goodmark, 2012; Meloy & Miller, 2011). Women in the United States may engage their informal networks first, then seek help from both informal networks and formal social service systems, and then, when the violence is most severe, turn to the legal system. Longitudinal research that explores whether women change their pattern over time or stay consistent and, if changes occur over time, what variables are associated with such change, would be a valuable contribution moving forward.

Importantly, the current analysis identified distinct demographic factors associated with help-seeking patterns in response to IPV. Consistent with Durfee and Messing (2012), this analysis indicated that Black women had a higher probability of being in the Broadly Engages Networks but Avoids Legal Systems class as opposed to the Broadly Engages Formal and Informal Networks class. These findings may reflect perceptions on the part of women of color that the legal system is biased or less helpful than other alternatives (Richie, 2012). Engaging the legal system may also imply a sense of betrayal of communities of color (Gómez & Freyd, 2018), thus making Black women hesitant to report abusive partners to the police or pursue criminal charges in an effort to enhance their own safety. However, the similar class found by Hanson et al. (2019; Avoiding the Justice System), did not indicate differences in race/ethnicity. In future studies, it may be helpful to explore the perceived effectiveness and barriers to using the legal system in addressing IPV and account for current events such as class action lawsuits for brutality, highly publicized events related to shooting citizens, or harassment of people of color.

Similar to existing studies that highlight the role of children in women’s decision making to seek help (Durfee & Messing, 2012; Duterte et al., 2008; Henning & Klesges, 2002; Randell et al., 2012), women in this study who engaged broadly with networks and services had more minor children compared to women who primarily engaged only informal networks. Women with children under 18 years old may perceive a greater need to address IPV through a combination of formal and informal networks, perhaps due to concerns about the wellbeing of their children. This finding points to the need for more comprehensive services that can simultaneously address women and their children’s needs (e.g., witnessing IPV, temporary housing). Also consistent with previous findings (Cho et al., 2020; Duterte et al., 2008; Flicker et al., 2011; Hanson et al., 2019; Henning & Klesges, 2002; Leone et al., 2007; Macy et al., 2005), the current analysis revealed an association between higher levels of IPV severity, mental health, and women’s broad engagement of networks and services. These findings provide support for theories that posit survivors develop strategies and expand actions to enhance their own safety when faced with increasing levels of violence (Goodman et al., 2003; Hamby & Gray-Little, 1997, 2007). These results also echo theories of help seeking explaining processes in which women’s appraisal of their abuse and perceived needs shape their decision making (Liang et al., 2005); in this study, symptoms of depression and PTSD may constitute needs perceived by women and thus facilitate help seeking.

Although employment status, education level, household income, age, abuse duration, and physical health were associated with women’s help seeking in previous studies (e.g., Cho et al., 2020; Durfee & Messing, 2012; Flicker et al., 2011; Henning & Klesges, 2002; Leone et al., 2007; Macy et al., 2005), our analysis did not detect any significant associations between those factors and class membership. The analysis, however, did reveal that foreign-born survivors were more likely to be in the Broadly Engages Formal and Informal Networks class than in the Broadly Engages Networks but Avoids Legal Systems class. This finding does not align with evidence of the immense barriers that immigrant IPV survivors face in accessing formal services (e.g., Raj & Silverman, 2002; Vidales, 2010). Eligibility criteria, which required study participants to speak and read English, may provide a possible explanation as the subsample of foreign-born survivors may have been more likely to be documented and in the United States for a longer period of time. Research has illuminated how language, immigration status, and time in the United States can influence risk and impede survivors’ efforts to seek help and engage domestic violence services (Messing et al., 2013; Raj & Silverman, 2002; Vidales, 2010; Wachter et al., 2019). There is a need for targeted research to further study patterns of IPV help seeking among diverse samples of immigrants.

Implications for Practice

The findings offer insights into opportunities and priorities for practice. Domestic violence practitioners should be aware of the multiple and varied ways that women choose or avoid seeking help. Understanding that women who are seeking services are looking for additional avenues of support may assist advocates in providing appropriate referrals and attending to the specific needs of individuals and their families. Practitioners should also explore women’s preferences and reluctances in engaging various resources so that providers can increase the range of women’s help-seeking options. Findings from this analysis highlight the salience of informal support in help-seeking processes and support the need for interventions that engage survivors’ informal networks as a part of domestic violence outreach and response services (Goodman & Smyth, 2011; Wachter et al., in press). Broad dissemination of information about service options and measures to address IPV is a matter of public health and safety. As indicated by a study, when surveying online apps for violence prevention, the capabilities range from education, documenting evidence for future cases, to instantaneous notice during an emergency (Sinha et al., 2019). Incorporating technology and innovative bystander intervention models thus may be another way of reaching survivors who rely on informal networks (Emezue, 2020; Glass et al., 2017; Lindsay et al., 2013; Renzetti & Follingstad, 2017). The public is becoming more involved in discussions around sexual violence, feminism, and equality, making the dissemination of online tools more feasible than before (Storer & Rodriguez, 2020). In order to provide more immediate help to survivors, it is crucial to maximize the opportunities afforded through new technologies to help provide quality resources to the informal networks they so often rely on.

Continued advocacy for the adoption and scale-up of culturally sensitive and trauma-informed practices is necessary to address barriers preventing women from accessing supportive services. Results indicating that women with higher severity of IPV, mental health problems, and more children engage more systems reiterates the need for comprehensive services and strong collaborative networks of providers. They also highlight the importance of early detection and intervention as service needs become more complex and multi-faceted over time (Messing et al., 2015). This study highlighted the extent to which women’s help-seeking strategies may include seeking help for an abusive partner (66% in this sample). Because people who use violence in their relationships are ultimately responsible for changing their behavior, this finding emphasizes the need to expand evidence-based, effective service options for people who use violence against intimate partners. A recent meta-analysis showed that rigorous evidence (e.g., randomized controlled trials) is lacking for existing batterer intervention programs (Cheng et al., 2019). Finally, the analysis reinforces the urgency of addressing structural factors that impede women—particularly women of color—from engaging legal systems and broad networks of potential support. More advocacy with police in appropriately responding to IPV are needed (Decker et al., 2020; Meloy & Miller, 2011).

Study Limitations

This study has several limitations to consider. First, the current analysis employed a non-representative sample, which limits the generalizability of the results to other service-engaged women and does not support transferability to survivors who do not seek help from and access the service systems. Second, the cross-sectional retrospective design does not allow for understanding the changes in patterns of women’s help seeking over time. Third, the current study only utilized women’s responses to items specifically related to help seeking on the Intimate Partner Violence Strategies Index (Goodkind et al., 2004; Goodman et al., 2003) but did not use strategies such as placating or resisting a partner that are not specific to help seeking. This limited exploration of all possible strategies that survivors used to manage the violence that they faced, but was consistent with our research questions which addressed help seeking. Fourth, the current study did not look at the outcomes of women’s actions and strategies (such as whether this action was helpful to address IPV), which is a gap in the IPV research literature and an important area of inquiry to prioritize moving forward. Last, when investigating the relationship between LCA class membership and demographics, relationship characteristics, and health factors, the current study did not conduct multivariate analysis but only focused on bivariate relationship. It is possible that some associations may be weaker if using multivariate analysis, warranting further examination with a bigger sample size.

Conclusions

Findings from this article show the pressing need for comprehensive and collaborative service networks that account for the salience of survivors’ informal networks, effectively engage men who use violence and remove barriers to accessing services from legal systems and social services. It is imperative to address structural factors limiting the range of formal help-seeking options available to survivors of IPV, particularly women of color. In collaboration with service providers, future research should prioritize longitudinal research with service-engaged survivors, as well as those who are not accessing services, to observe women’s help seeking over time.

Footnotes

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship and/or publication of this article: Opinions or points of view expressed in this document are those of the authors and do not necessarily reflect the official position of, or a position that is endorsed by, CNCS.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This material is based upon work supported by the Corporation for National & Community Service (CNCS) under Grant #18AFHAZ0010009.

ORCID iDs

Shih-Ying Cheng

Karin Wachter

Andrea Kappas

Jill Theresa Messing

Meredith Bagwell-Gray

Author Biographies

Shih-Ying Cheng, MSW, is a doctoral candidate in the School of Social Work at Washington University in St. Louis. Her research interests include services and policies that address intimate partner violence, health and social service utilization among recent immigrants and the implementation of evidence-based practice.

Karin Wachter, MEd, PhD, is an assistant professor in the School of Social Work at Arizona State University. Her background is in international humanitarian assistance, and her research currently focuses on issues of violence against women and social support in forced migration.

Andrea Kappas, MSW, is a doctoral candidate in the School of Social Work at Arizona State University. Her research focuses on examining trauma-informed complementary and alternative approaches to treatment among individuals with experiences of interpersonal victimization, with an emphasis on intimate partner violence.

Megan Lindsay Brown, MSW, PhD, is a research assistant professor at Arizona State University in the School of Social Work. Her research focuses on the use of information communication technologies (ICT) within intimate relationships, and especially technology-based abuse. Applied projects include adapting technology to support victims of domestic violence.

Jill Theresa Messing, MSW, PhD, is a professor in the School of Social Work and the Director of the Office of Gender-Based Violence at Arizona State University. Dr. Messing specializes in intimate partner violence risk assessment. She has evaluated the predictive validity of multiple intimate partner violence risk assessments including several forms of the Danger Assessment. As a social worker, Dr. Messing is particularly interested in the use of risk assessment in evidence-based practice and with the development and testing of innovative interventions for victims of intimate partner violence.

Meredith Bagwell-Gray, MSSW, PhD, is an assistant professor in the School of Social Welfare at The University of Kansas. Her research centers on designing, implementing, and testing multilevel interventions to improve the health and safety of survivors of intimate partner violence. She approaches this work from a critical perspective, using an intersectional framework to examine oppression and resistance, as well as a strengths perspective that emphasizes post-traumatic growth and resiliency.

Tina Jiwatram-Negron, MSW, PhD, is an assistant professor in the School of Social Work at Arizona State University. Her research is focused on examining and addressing IPV/GBV and associated health and mental health disparities among socially vulnerable and marginalized women domestically and internationally.

References

Anyikwa

V. A.

(2015). The intersections of race and gender in help-seeking strategies among a battered sample of low-income African American women. Journal of Human Behavior in the Social Environment , 25(8), 948–959.

Asparouhov

, & Muthén

(2018). Auxiliary variables in mixture modeling: Using the BCH method in Mplus to estimate a distal outcome model and an arbitrary secondary model. Mplus Web Notes 21 . http://statmodel.com/examples/webnotes/webnote21.pdf

Bakk

, Tekle

F. B.

, & Vermunt

J. K.

(2013). Estimating the association between latent class membership and external variables using bias-adjusted three-step approaches. Sociological Methodology , 43(1), 272–311.

Bakk

, & Vermunt

J. K.

(2016). Robustness of stepwise latent class modeling with continuous distal outcomes. Structural equation modeling: A multidisciplinary Journal , 23(1), 20–31.

Ben-Porat

(2020). Patterns of service utilization among women who are victims of domestic violence: The contribution of cultural background, characteristics of violence, and psychological distress. Journal of Interpersonal Violence , 35(17–18), 3167–3187. https://doi.org/10.1177/0886260517707308

Bergman

L. R.

, & Trost

(2006). The person-oriented versus the variable-oriented approach: Are they complementary, opposites, or exploring different worlds? Merrill-Palmer Quarterly , 52(3), 601–632.

Bolck

, Croon

, & Hagenaars

(2004). Estimating latent structure models with categorical variables: One-step versus three-step estimators. Political Analysis , 12, 3–27. https://doi.org/10.1093/pan/mph001

Cheng

S.-Y.

, Davis

, Jonson-Reid

, & Yaeger

(2019). Compared to what? A meta-analysis of batterer intervention studies using nontreated controls or comparisons. Trauma, Violence, & Abuse . https://doi.org/10.1177/1524838019865927

Cho

, Shamrova

, Han

J.-B.

, & Levchenko

(2020). Patterns of intimate partner violence victimization and survivors’ help-seeking. Journal of Interpersonal Violence , 35(21–22), 4558–4582. https://doi.org/10.1177/0886260517715027

10.

Coker

A. L.

, Smith

P. H.

, Thompson

M. P.

, McKeown

R. E.

, Bethea

, & Davis

K. E.

(2002). Social support protects against the negative effects of partner violence on mental health. Journal of Women’s Health & Gender-based Medicine , 11(5), 465–476.

11.

Collins

L. M.

, & Lanza

S. T.

(2010). Latent class and latent transition analysis: With applications in the social, behavioral, and health sciences . John Wiley & Sons.

12.

Conybeare

, Behar

, Solomon

, Newman

M. G.

, & Borkovec

T. D.

(2012). The PTSD Checklist-Civilian Version: Reliability, validity, and factor structure in a nonclinical sample. Journal of Clinical Psychology , 68(6), 699–713. https://doi.org/10.1002/jclp.21845

13.

Crenshaw

K. W.

(2017). On intersectionality: Essential writings . The New Press.

14.

Davies

, & Lyon

(2014). Domestic violence advocacy: Complex lives/difficult choices (2nd ed.). Sage Publications.

15.

Decker

M. R.

, Holliday

C. N.

, Hameeduddin

, Shah

, Miller

, Dantzler

, & Goodmark

(2020). Defining justice: Restorative and retributive justice goals among intimate partner violence survivors. Journal of Interpersonal Violence . https://doi.org/10.1177/0886260520943728

16.

Durfee

, & Messing

J. T.

(2012). Characteristics related to protection order use among victims of intimate partner violence. Violence Against Women , 18(6), 701–710.

17.

Duterte

E. E.

, Bonomi

A. E.

, Kernic

M. A.

, Schiff

M. A.

, Thompson

R. S.

, & Rivara

F. P.

(2008). Correlates of medical and legal help seeking among women reporting intimate partner violence. Journal of Women’s Health , 17(1), 85–95.

18.

Dutton

M. A.

(1992). Empowering and healing the battered woman: A model for assessment and intervention . Springer Publishing Company.

19.

Eaton

W. W.

, Smith

, Ybarra

, Muntaner

, & Tien

(2004). Center for Epidemiologic Studies Depression Scale: Review and revision (CESD and CESD-R). In Maruish

M. E.

(Ed.), The use of psychological testing for treatment planning and outcomes assessment: Instruments for adults (pp. 363–377). Routledge.

20.

Eden

K. B.

, Perrin

N. A.

, Hanson

G. C.

, Messing

J. T.

, Bloom

T. L.

, Campbell

J. C.

, Gielen

A. C.

, Clough

A. S.

, Barnes-Hoyt

J. S.

, & Glass

N. E.

(2015). Use of online safety decision aid by abused women: Effect on decisional conflict in a randomized controlled trial. American Journal of Preventive Medicine , 48(4), 372–383.

21.

El-Khoury

M. Y.

, Dutton

M. A.

, Goodman

L. A.

, Engel

, Belamaric

R. J.

, & Murphy

(2004). Ethnic differences in battered women’s formal help-seeking strategies: A focus on health, mental health, and spirituality. Cultural Diversity and Ethnic Minority Psychology , 10(4), 383.

22.

Emezue

(2020). Digital or digitally delivered responses to domestic and intimate partner violence during COVID-19. JMIR . https://doi.org/10.2196/preprints.19831

23.

Fanslow

J. L.

, & Robinson

E. M.

(2010). Help-seeking behaviors and reasons for help seeking reported by a representative sample of women victims of intimate partner violence in New Zealand. Journal of Interpersonal Violence , 25(5), 929–951.

24.

Flicker

S. M.

, Cerulli

, Zhao

, Tang

, Watts

, Xia

, & Talbot

N. L.

(2011). Concomitant forms of abuse and help-seeking behavior among White, African American, and Latina women who experience intimate partner violence. Violence Against Women , 17(8), 1067–1085.

25.

Glass

, Clough

, Case

, Hanson

, Barnes-Hoyt

, Waterbury

, Alhusen

, Ehrensaft

, Grace

K. T.

, & Perrin

(2015). A safety app to respond to dating violence for college women and their friends: The MyPlan study randomized controlled trial protocol. BMC Public Health , 15(1), 871.

26.

Glass

, Perrin

N. A.

, Hanson

G. C.

, Bloom

T. L.

, Messing

J. T.

, Clough

A. S.

, Campbell

J. C.

, Gielen

A.C.

, Case

, & Eden

K. B.

(2017). The longitudinal impact of an internet safety decision aid for abused women. American Journal of Preventive Medicine , 52(5), 606–615.

27.

Gómez

J. M.

, & Freyd

J. J.

(2018). Psychological outcomes of within-group sexual violence: Evidence of cultural betrayal. Journal of Immigrant and Minority Health , 20(6), 1458–1467.

28.

Gondolf

E. W.

, & Fisher

E. R.

(1988). Battered women as survivors: An alternative to treating learned helplessness . Lexington Books.

29.

Goodkind

J. R.

, Sullivan

C. M.

, & Bybee

D. I.

(2004). A contextual analysis of battered women’s safety planning. Violence Against Women , 10(5), 514–533.

30.

Goodman

, Dutton

M. A.

, Weinfurt

, & Cook

(2003). The intimate partner violence strategies index: Development and application. Violence Against Women , 9(2), 163–186.

31.

Goodman

, & Smyth

K. F.

(2011). A call for a social network-oriented approach to services for survivors of intimate partner violence. Psychology of Violence , 1(2), 79.

32.

Goodmark

(2012). A troubled marriage: Domestic violence and the legal system . New York University Press.

33.

Hamby

(2014). Battered women’s protective strategies: Stronger than you know . Oxford University Press.

34.

Hamby

, & Gray-Little

(1997). Responses to partner violence: Moving away from deficit models. Journal of Family Psychology , 11(3), 339.

35.

Hamby

, & Gray-Little

(2007). Can battered women cope? A critical analysis of research on women’s responses to violence. In Kendall Tackett

& Giacomoni

(Eds.), Intimate partner violence (pp. 28–21). Civic Research Institute.

36.

Hanson

G. C.

, Messing

J. T.

, Anderson

J. C.

, Thaller

, Perrin

N. A.

, & Glass

N. E.

(2019). Patterns and usefulness of safety behaviors among community-based women survivors of intimate partner violence. Journal of Interpersonal Violence . https://doi.org/0886260519853401

37.

Hegarty

, & Valpied

(2013). Composite Abuse Scale Manual . University of Melbourne.

38.

Henning

K. R.

, & Klesges

L. M.

(2002). Utilization of counseling and supportive services by female victims of domestic abuse. Violence and Victims , 17(5), 623.

39.

Johnson

M. P.

, & Leone

J. M.

(2005). The differential effects of intimate terrorism and situational couple violence: Findings from the National Violence Against Women Survey. Journal of Family Issues , 26(3), 322–349. https://doi.org/10.1177/0192513x04270345

40.

Leone

J. M.

, Johnson

M. P.

, & Cohan

C. L.

(2007). Victim help seeking: Differences between intimate terrorism and situational couple violence. Family Relations , 56(5), 427–439.

41.

Liang

, Goodman

, Tummala-Narra

, & Weintraub

(2005). A theoretical framework for understanding help-seeking processes among survivors of intimate partner violence. American Journal of Community Psychology , 36(1–2), 71–84.

42.

Lindsay

, Messing

J. T.

, Thaller

, Baldwin

, Clough

, Bloom

, Eden

K. B.

, & Glass

(2013). Survivor feedback on a safety decision aid smartphone application for college-age women in abusive relationships. Journal of Technology in Human Services , 31(4), 368–388.

43.

Lipsky

, Caetano

, Field

C. A.

, & Larkin

G. L.

(2006). The role of intimate partner violence, race, and ethnicity in help-seeking behaviors. Ethnicity and Health , 11(1), 81–100.

44.

Macy

R. J.

, Nurius

P. S.

, Kernic

M. A.

, & Holt

V. L.

(2005). Battered women’s profiles associated with service help-seeking efforts: Illuminating opportunities for intervention. Social Work Research , 29(3), 137–150.

45.

Meloy

M. L.

, & Miller

S. L.

(2011). The victimization of women: Law, policies, and politics . Oxford University Press.

46.

Messing

J. T.

, Amanor-Boadu

, Cavanaugh

C. E.

, Glass

N. E.

, & Campbell

J. C.

(2013). Culturally competent intimate partner violence risk assessment: Adapting the danger assessment for immigrant women. Social Work Research , 37(3), 263–275.

47.

Messing

J. T.

, Ward-Lasher

, Thaller

, & Bagwell-Gray

M. E.

(2015). The state of intimate partner violence intervention: Progress and continuing challenges. Social Work , 60(4), 305–313.

48.

Meyer

(2010). Seeking help to protect the children?: The influence of children on women’s decisions to seek help when experiencing intimate partner violence. Journal of Family Violence , 25(8), 713–725.

49.

Muthén

L. K.

, & Muthén

(1998–2017). Mplus (8th ed.). Muthén & Muthén.

50.

Nurius

P. S.

, Macy

R. J.

, Nwabuzor

, & Holt

V. L.

(2011). Intimate partner survivors’ help-seeking and protection efforts: A person-oriented analysis. Journal of Interpersonal Violence , 26(3), 539–566.

51.

Nylund

K. L.

, Asparouhov

, & Muthén

B. O.

(2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural equation modeling: A multidisciplinary Journal , 14(4), 535–569.

52.

Overstreet

N. M.

, & Quinn

D. M.

(2013). The intimate partner violence stigmatization model and barriers to help seeking. Basic and Applied Social Psychology , 35(1), 109122.

53.

Raj

, & Silverman

(2002). Violence against immigrant women: The roles of culture, context, and legal immigrant status on intimate partner violence. Violence Against Women , 8(3), 367–398.

54.

Ramaswamy

, Desarbo

W. S.

, Reibstein

D. J.

, & Robinson

W. T.

(1993). An empirical pooling approach for estimating marketing mix elasticities with PIMS data. Marketing Science , 12(1), 103–124.

55.

Randell

K. A.

, Bledsoe

L. K.

, Shroff

P. L.

, & Pierce

M. C.

(2012). Mothers’ motivations for intimate partner violence help-seeking. Journal of Family Violence , 27(1), 55–62.

56.

Renzetti

, & Follingstad

(2017). Preventing intimate partner violence: Interdisciplinary perspectives . Policy Press.

57.

Richie

B. E.

(2012). Arrested justice: Black women, violence, and America’s prison nation . New York University Press.

58.

Shannon

, Logan

, Cole

, & Medley

(2006). Help-seeking and coping strategies for intimate partner violence in rural and urban women. Violence and Victims , 21(2), 167–181.

59.

Sinha

, Shrivastava

, & Paradis

(2019). A survey of the mobile phone-based interventions for violence prevention among women. Advances in Social Work , 19(2), 493–517.

60.

Smith

P. H.

, Earp

J. A.

, & DeVellis

(1995). Measuring battering: Development of the Women’s Experience with Battering (WEB) Scale. Women’s Health: Research on Gender, Behavior, and Policy , 1(4), 273–288.

61.

Storer

H. L.

, & Rodriguez

(2020). #Mapping a movement: Social media, feminist hashtags, and movement building in the digital age. Journal of Community Practice , 28(2), 160–176.

62.

US Department of Veteran Affairs. (2012). Using the PTSD checklist (PCL) . National Center for PTSD.

63.

Vatnar

S. K. B.

, & Bjørkly

(2009). Interactional aspects of intimate partner violence result in different help-seeking behaviors in a representative sample of women. Journal of Family Violence , 24(4), 231–241.

64.

Vermunt

J. K.

(2017). Latent class modeling with covariates: Two improved three-step approaches. Political Analysis , 18(4), 450–469. https://doi.org/10.1093/pan/mpq025

65.

Vidales

G. T.

(2010). Arrested justice: The multifaceted plight of immigrant Latinas who faced domestic violence. Journal of Family Violence , 25(6), 533–544.

66.

Wachter

, Cook Heffron

, Dalpe

, & Spitz

(in press). “Where is the women’s center here?”: The role of information in refugee women’s help seeking for intimate partner violence in a resettlement context. Violence Against Women .

67.

Wachter

, Dalpe

, & Heffron

L. C.

(2019). Conceptualizations of domestic violence–related needs among women who resettled to the United States as refugees. Social Work Research , 43(4), 207–219.

68.

Walker

L. E. A.

(1979). The battered women . Harper & Row Publishers.

69.

Wang

, & Wang

(2012). Structural equation modeling: Applications using Mplus. John Wiley & Sons.