Abstract
Knowledge about where battered women present for services and the violence, biopsychosocial, and demographic factors associated with their help seeking can provide social workers with guidance in anticipating needs among this portion of their clientele. The authors examined the service contact patterns of a sample of battered women (N = 448) following an incident of partner violence that triggered legal involvement. Significant group differences, tested with t tests and chi squares, between women who sought compared with those who did not seek services were found on partner violence exposure and biopsychosocial factors. Correlations and regression analyses of relationships among partner violence and biopsychosocial and demographic factors with help-seeking indices show how battered women’s needs differentially relate across a range of service types. Results show distinctive profiles of needs and resources among battered women who seek violence, legal, health, economic, substance abuse, and religious helping services.
Keywords: battered women, depression, health, help seeking, partner violence, social support, substance abuse
Although they may not be aware of it, social workers see battered and endangered women in their practices every day. Battered women frequently seek support from their social networks and human services providers for a range of life problems without calling attention to the violence in their lives (Henning & Klesges, 2002). Comparatively few battered women seek services from providers who specialize in domestic violence. Social workers span a broad spectrum of human services agencies and are well positioned to facilitate abused women’s entry into service systems equipped to address the violence in their lives (Nurius & Asplund, 1994).
If most women in violent relationships are not seeking domestic violence services, where are they going for help? What are their needs? What can we learn from violence exposure, biopsychosocial, and demographic factors to better anticipate their needs across diverse services? We addressed these questions through an investigation of services help-seeking efforts in a sample of battered women following an incident of intimate partner violence. Our goal is to provide practitioners with detailed information about the needs, resources, and characteristics of battered women as they seek help, predominantly from service providers other than violence specialists.
Research generally indicates that women cope with partner violence through informal means as long as possible (Coker, Derrick, Lumpkin, Aldrich, & Oldendick, 2000). As a result, battered women who seek formal services from agencies and providers may have long-standing needs and severe problems (Hutchison & Hirschel, 1998). Partner violence has a radiating impact, with “first order” effects on a woman’s physical and mental health and secondary effects on her ability to function in her social world (for example, school, work, or parenting), spanning out to affect others such as family, friends, and coworkers (Riger, Raja, & Camacho, 2002). Service needs illustrated through earlier research include help for biopsychosocial problems related to mental and physical health and injury (Golding, 1999; Plichta & Falik, 2001), substance use (El-Bassel, Gilbert, Schilling, & Wada, 2000), and employment and financial difficulties (Lloyd, 1997).
Although safety is a universal concern, battered women vary in the needs, characteristics, and resources that shape their experience of violence and help seeking (Mitchell & Hodson, 1986). On one hand, they may need formal services to manage the difficulties in their lives to escape or end the abuse. On the other hand, disclosing the abuse or trying to protect themselves and their children may increase their danger in terms of escalated or more insidious abuse, interference with work or schooling, damaged social relations, impairment of the woman’s parenting, or danger to children’s well-being (Wolf, Ly, Hobart, & Kernic, 2003). Inroads have been made in assessing battered women’s needs and resources that galvanize help-seeking efforts. In summarizing their own and others’ research, Hutchison and Hirschel (1998) found considerable variation in help seeking, with only a small proportion of women not seeking help eventually from sources outside their social network during the course of an abusive relationship. This and related work have provided initial findings linking demographic characteristics to services help seeking. However, research generally has not addressed biopsychosocial problems that might co-occur with partner violence. We addressed this gap through an examination of partner violence exposure, biopsychosocial, and demographic factors in relation to battered women’s help-seeking efforts across a range of human services directly following an incident of intimate partner violence.
We hypothesized that the combination of partner violence, biopsychosocial, and demographic factors would better profile battered women’s needs and resources related to human services contacts than would any one of these sets of factors alone. Consistent with prior research, we investigated three subtypes of partner violence—psychological, physical, and sexual abuse—to discern the differential relationships these three types of violence have on service help-seeking efforts (Sackett & Saunders, 2001). In addition to behavioral acts of partner violence, we examined women’s appraisals of the threat of their abusive partners (Smith, Earp, & DeVellis, 1995). We included biopsychosocial characteristics that have been found to be related to partner violence, including depression (Tjaden & Thoennes, 2000), physical illness and injuries (Petersen, Gazmarariam, & Andersen, 2001), and substance abuse (Levine & Greene, 2000). Social interactions and relationships are other biopsycho-social factors that have been found to be related to battered women’s help seeking (Barnett, Martinez, & Keyson, 1996). We examined positive and negative social relationships to determine if these qualitatively different kinds of social ties would have different implications for help seeking across different types of services (Lincoln, 2000). Finally, we investigated demographics to represent social structural factors, including employment, income, education, race/ethnicity, and age (Henning & Klesges, 2002).
Our goal was to help determine how battered women’s needs, characteristics, and resources are associated with help seeking across a spectrum of formal human services: domestic violence, legal, health care, public assistance, counseling, substance abuse, and religious or spiritual. We examined both successful and attempted but not successful help-seeking efforts, which allowed us to include participants’ help-seeking behaviors even if these efforts did not result in service delivery. For example, a woman may seek services from a domestic violence shelter but be turned away because the agency is full to capacity. First, we compared participants who reported attempts to or successful receipt of services help seeking with participants who reported no services help seeking. Subsequently, we analyzed patterns of needs and resources among participants who attempted to or obtained help across social, health, and legal services.
METHOD
Sample
Participants were adult (age 18 or older) female residents of Seattle, Washington, who were victims of abuse by a current or former male intimate partner (N = 448) that resulted in a police-reported incident or filing of a protection order in Seattle (see Wolf, Holt, Kernic, & Rivara, 2000). This abuse incident constituted the index episode that led to the individual’s recruitment. Of the 742 women eligible for the study, 124 (16.7%) refused to participate, 62 (8.4%) agreed but did not complete interviews, 108 (14.6%) could not be contacted, and 448 (60.4%) were enrolled. Participants and nonpartici-pants were similar on age, marital status, proportion having a child with the abuser, type of offense reported to the police, and proportion injured at the index incident. Participants were more likely than nonparticipants to have obtained protection orders (56.5% compared with 41.8%) and less likely to be living with the abuser at the time of the index incident (26.7% compared with 36.4%).
The University of Washington’s Human Subjects Review Committee approved study protocols. Eligible study participants were telephoned approximately one month after the index incident and asked to participate by completing a questionnaire. Participants were given the option of completing an interviewer-administered telephone interview or a self-administered mailed copy of the questionnaire. Eighty-one percent chose the telephone interview. There were no significant differences between the participants who chose the mailed questionnaire and those who chose the telephone survey. At the incident date, the distribution of participants’ ages was 18 to 70 (M = 32.01, SD = 9.46). In terms of race and ethnicity, 52.0% of the participants described themselves as European American, 19.7% as African American, 6.6% as Asian/ Pacific Islander, 4.3% as Native American/Alaska Native, 12.9% as Hispanic/Latina, and 4.5% as biracial or a combination of the race/ethnicity categories. Fewer than 5.0% of the women reported being a citizen of a country other than the United States. The distribution of education attained was: eighth grade or less (2.2%), some high school (9.2%), high school graduate or GED (24.6%), post-high school training but not college (7.8%), some college (37.5%), college graduate (15.4%), advanced or graduate degree (3.3%). Participants’ reported occupational status at the time of the incident was employed full-time (49.1%), employed part-time (12.3%), attending school (3.8%), both attending school and employed (9.0%), not employed outside the home (25.3%). Fewer than one percent of the sample reported either being self-employed or disabled. The distribution of household income was less than $15,000 per year (43.9%), $15,000–$19, 999 (11.8%), $20,000–$24, 999 (8.3%), $25,000–$34, 999 (14.5%), $35,000– $49,999 (12.0%), $50,000–$49,000 (4.9%), $70,000 and above (4.7%).
Measures
Data were collected through survey questionnaires that followed a structured format using several well-evaluated measures described in the following sections.
Help Seeking
Participants were asked about help-seeking efforts on 22 items, including domestic violence services, legal services, health care, economic assistance, substance abuse treatment, counseling services, and religious-spiritual services. These items were developed in conjunction with the research advisory board members, who were primarily domestic violence services providers and reflected the range of human services available to and used by battered women in the greater Seattle area. Help seeking from domestic violence services providers included shelter or housing, support groups, advocacy, and domestic violence counseling. Legal services included assistance with divorce, custody or separation, police charge, protection order, prosecution, crime victim’s compensation, and support enforcement or paternity action. Economic help seeking was operationalized in terms of receiving welfare, food stamps, or social security, and food bank services. Substance abuse treatment help seeking included drug treatment, alcohol treatment, and Alcoholics Anonymous or Narcotics Anonymous. General counseling services included private counseling, and Al-Anon. One item asked about help seeking with religious or spiritual organizations. Help seeking with health care focused on services as a result of the partner violence index incident only: receiving first aid at scene of incident from paramedics, seeking out a health care provider in an office or clinic, visiting an emergency room, and staying overnight in a hospital.
Participants were instructed to report on services help-seeking efforts in the time since the index event. Participants were asked if they received a specific service, attempted to obtain the service but did not receive the service, or did not attempt to obtain the service. Help seeking was coded as accessed = 2, attempted = 1, or no attempt = 0. In a specific service category (for example, all items related to legal services), help-seeking items for each participant were summed to create weighted help-seeking indices by service type. Health care services help-seeking responses were coded 0 for no and 1 for yes; the responses were summed. An aggregated help-seeking scale was created by summing the values of all types of help seeking.
Partner Violence
Behavioral indicators of partner violence were severity of physical, sexual, and psychological abuse in the year preceding the index incident and at the incident that triggered legal involvement, measured using the subscales from the revised Conflict Tactics Scale (CTS2) (Straus, Hamby, Boney-McCoy, & Sugarman, 1996). Following the CTS2 guidelines, respondents indicated the frequency of physical abuse for the preceding year (12 items summed; α = .90, M = 24.15, SD = 38.49), sexual abuse (two items summed; α = .59, M = 5.31, SD = 10.27), and psychological abuse (eight items summed; (α = .83, M = 60.08, SD = 44.81). For the index event, participants were asked to respond to the same items of abuse as used for the preceding year. However, rather than indicating a frequency for each type of abuse, participants indicated whether that type of abuse occurred at the incident (1 = yes, 0 = no), with responses summed to create the physical abuse scale (α = .85, M = 2.82, SD = 2.92), the sexual abuse scale (α = .41, M = .15, SD = .41), and the psychological abuse scale (α = .70, M = 3.85, SD = 1.90).
To examine threat of partner violence, we used the Women’s Experiences with Battering scale (WEB) (Smith et al., 1995), which measures victims’ psychological vulnerability through women’s perceptions of their susceptibility to physical and psychological danger, loss of power, and loss of control in a relationship with a male partner. The WEB has demonstrated high internal consistency, good construct validity, and significant correlation with known-group status (Smith et al., 1995). The 10 WEB items are rated on a six-point Likert scale (ranging from 1 = strongly disagree to 6 = strongly agree) and were summed to create a scale (α = .94, M = 39.35, SD = 16.44).
Biopsychosocial Factors
Social relationships were measured using the Social Adjustment Scale (SAS-M),which measures engagement with friends and extended family members and participation in leisure activities (Cooper, Osborn, Gath, & Feggetter, 1982; Weissman & Paykel, 1974). Cooper and colleagues adapted the original scale to use a five-point Likert scale (ranging from 1 = not at all to 5 = all the time) to rate all items. Items were clustered into positive (seven items, for example, “Been able to talk about your feelings openly with your friends,” “Depended on your relatives for help, advice, or friendship”) and negative sets (nine items, for example, “Gotten angry with or argued with your friends,” “Been feeling that your relatives had let you down”). Items for each set were averaged into a positive social relationships/activities scale (α = .80, M = 3.25, SD = .98) and a negative social relationships/activities scale (α = .69, M = 2.00, SD = .67).
Depression symptomatology was measured with the Center for Epidemiologic Studies Depression Scale (CES-D) items for the one-week period following the partner violence index incident. The 20 items rated on a four-point Likert scale (ranging from 0 = rarely or none of the time to 3 = most or all of the time) were scored following standard protocols (α = .95, M = 32.44, SD = 14.74). A score of 16 or less on the CES-D indicates no depression, mild depression is indicated by scores of 17 through 26, and severe depression is indicated when scores are 27 or higher.
The standardized summary score for the physical health component short-form 12 (PCS-12) (Ware, Kosinski, & Keller, 1995), a shortened version of the physical health component 36 comprised of eight health concepts, was used to measure the women’s overall physical functioning level (M = 46.66, SD = 10.68). The reliability and validity of this measure has been extensively investigated and discussed elsewhere (Ware et al.). Indicator variables of 0 and 1 were created from the response choice categories and weighted using regression coefficients from the general U.S. population and aggregated, allowing comparison of the results across studies. Among U.S. women, the mean PCS-12 score is 49.11 with a standard deviation of 9.92; lower scores on the PCS-12 are indicative of poorer physical health.
To measure injuries at incident, participants were asked to indicate if they received the following types of injuries: pain; bruises and/or soreness or swelling; cuts and/or bleeding; broken bone(s); internal injury(ies); loss of consciousness. The positive responses to each injury item were summed to provide a summary injury score.
Alcohol abuse was measured using three items from the NET, a commonly used alcohol abuse screening tool (Russell et al., 1994). Participants reported on alcohol abuse for the year before the incident; items included “have a drink first thing in the morning to steady your nerves or get rid of a hangover,” “feel you ought to cut down on your drinking,” and “awaken in the morning after drinking the night before and find that you could not remember a part of the evening.”
Drug abuse was measured by one item, which asked if the participant had used illegal drugs in the year before the incident, coded 0 for no and 1 for yes.
Demographic Factors
Income, education, and employment were assessed ordinally with ascending values: 1 to 7 for income the year preceding the index event (less than $15,000, $15,000–$19,999, $20,000–$24,999, $25,000–$34,999, $35,000– $49,999, $50,000–$69,999, $70,000 and over); 1 to 7 for education (eighth grade or less, some high school completion, high school graduate/GED, post-high school vocational training other than college, some college, college graduate, and advanced or graduate degree), and 0 to 2 for employment/ occupation at the time of the index event (was not employed outside the home; was employed part-time or was a student; and was employed outside the home full-time). Employment is an important demographic indicator to investigate in battered women, because it represents access to supports, contacts, and information outside of the home as well as a critical economic resource. A woman may have lower income as a student or part-time employee relative to some full-time employed women. Still, these statuses indicate movement toward increased earning potential and access to contacts or supports outside of her home, which may facilitate and galvanize help seeking. Full-time employment was given the highest value because it represents access to supports and contacts outside of home in addition to its economic value.
Racial and ethnic groups were first examined separately through the use of a one-way ANOVA on variables of interest; no significant differences were found among the women of color on the help-seeking indices. As a result, all women of color (n = 230) were coded 1 and compared with the European American women (n = 212) coded 0” Involvement with the perpetrator was assessed through the question “Were you intimately involved or in a dating relationship” with the perpetrator on the incident date. Participants who indicated yes (n = 244) were given the code of 1, and those who indicated no (n = 204) were given the code of 0, indicating that participants were formerly involved with the perpetrator at the time of the incident (for example, an ex-husband).
Analysis
Participants were first divided into services help seekers, including women who attempted to obtain or did obtain services, and women who reported no help seeking. Group differences were tested through t tests and chi squares to determine if there were significant violence exposure, biopsychosocial, and demographic differences between the participants who did and did not seek services. Second, among the participants who sought or obtained services, bivariate correlations were used to determine the relationships of battered women’s characteristics to the services they sought. Multivariate forward stepwise regression analyses, using p < .05 and p < .10 as the entrance and removal criterion, respectively, were then conducted by regressing the partner violence, biopsychosocial, and demographic factors on the help-seeking indices and the aggregated help-seeking index to determine the most efficient set of multivariate needs and resources related to help seeking. The coefficients, the R2, and the overall F for each step were examined to determine the significance of each step. The stepwise regression analyses included only the women who reported attempting to obtain or obtaining formal help to determine the needs and resources related to services help seeking. All analyses were conducted in SPSS 11.0.
RESULTS
In the aggregated index 85% of participants reported having engaged in some type and level of service help seeking. The type of service most sought was legal (66%), with more than one-third of the women reporting contact with domestic violence (38%), economic (32%), and counseling services (31%) (Table 1).
Table 1.
Percentages, Means, and Ranges of Participant Help-Seeking Efforts across Service Categories
| Participant Help-Seeking |
Domestic Violence |
Legal | Health | Economic | Substance Treatment |
Counseling | Religious- Spiritual |
Aggregated |
|---|---|---|---|---|---|---|---|---|
| Number of service items | 4 | 6 | 4 | 2 | 3 | 2 | 1 | 22 |
| Percent help seeking | 38 | 66 | 17 | 32 | 9 | 31 | 7 | 85 |
| Mean | 1.25 | 2.68 | .27 | .70 | .25 | .60 | .13 | 5.88 |
| Range | 0–8 | 0–12 | 0–4 | 0–4 | 0–6 | 0–4 | 0–2 | 0–29 |
In the t test and chi-square analyses, service help-seeking participants experienced significantly higher rates of partner violence in the preceding year and at the index incident, greater threat appraisals, and greater biopsychosocial needs, including higher levels of depression and negative social relationships and lower levels of physical health functioning (Table 2). By contrast, service help-seeking participants did not differ in their demographic characteristics from the participants who reported no service help-seeking efforts.
Table 2.
Group Differences between Participants Who Reported Service Help-Seeking Efforts and Participants Who Reported No Service Help-Seeking Efforts
| Non-Help Seekers (N = 66) |
Help Seekers (N = 382) |
|||||
|---|---|---|---|---|---|---|
| Variable | df | t or χ2 | M or % | SD | M or % | SD |
| Partner violence | ||||||
| Year psychological abuse | 446 | 3.38*** | 43.08 | 36.71 | 63.02 | 45.46 |
| Year physical abuse | 445 | 2.95** | 11.35 | 29.72 | 26.36 | 39.42 |
| Year sexual abuse | 446 | 2.92** | 1.94 | 6.53 | 5.89 | 10.67 |
| Incident psychological abuse | 446 | 2.06* | 3.41 | 1.84 | 3.93 | 1.91 |
| Incident physical abuse | 446 | 2.43* | 2.02 | 2.44 | 2.96 | 2.98 |
| Incident sex abuse | 446 | .60 | .12 | .37 | .15 | .42 |
| Threat appraisals | 446 | 4.80*** | 30.59 | 16.90 | 40.87 | 15.90 |
| Biopsychosocial | ||||||
| Positive social activities | 446 | 1.13 | 3.12 | 1.01 | 3.27 | .99 |
| Negative social activities | 446 | 3.80*** | 1.73 | .57 | 2.06 | .67 |
| Depressive symptoms | 446 | 4.66*** | 24.80 | 15.31 | 33.75 | 14.24 |
| Physical health functioning | 446 | 2.94** | 50.31 | 9.96 | 46.05 | 10.69 |
| Alcohol abuse | 430 | .95 | .76 | 1.24 | .62 | 1.04 |
| Injuries at incident | 446 | 1.87a | .91 | 1.12 | 1.24 | 1.35 |
| Involved with perpetrator | 1 | χ2 = .06 | yes: 53% | yes: 55% | ||
| no: 47% | no: 45% | |||||
| Substance use | 1 | χ2 = .06 | yes: 22% | yes: 20% | ||
| no: 78% | no: 80% | |||||
| Demographics | ||||||
| Income | 406 | 1.59 | 3.08 | 1.91 | 2.66 | 1.90 |
| Education | 446 | .38 | 4.23 | 1.30 | 4.30 | 1.45 |
| Employment | 444 | 1.49 | 1.39 | .78 | 1.21 | .84 |
| Age | 446 | 1.32 | 33.42 | 10.80 | 31.77 | 9.20 |
| European American/women of color | 1 | χ2 = .80 | European America n: 47% women of color: 53% |
European America n:53% women of color:47% |
||
Approaches statistical significance at p ≤ .06.
p ≤ .05.
p ≤ .01.
p ≤ .001.
The correlations show relationships among service help-seeking indices with partner violence, biopsychosocial, and demographic factors; domestic violence service help-seeking efforts significantly and positively related to all other categories of help seeking except for economic services (Table 3). Domestic violence help seeking also significantly and positively related to past-year and incident psychological, physical, and sexual violence and women’s threat appraisals of their partners, although not to sexual abuse at incident. Domestic violence service help seeking was significantly and positively associated with all biopsychosocial variables except drug use and alcohol abuse. Age was the only demographic variable significantly associated with domestic violence services, with age positively associated with these help-seeking efforts.
Table 3.
Correlations among Service Help-Seeking Efforts Indices and with Partner Violence Exposure, Biopsychosocial, and Demographic Factors (N = 382)
| Factor | Domestic Violence |
Legal | Health | Economic | Substance Treatment |
Counseling | Religious- Spiritual |
|---|---|---|---|---|---|---|---|
| Help seeking | |||||||
| Domestic violence | — | ||||||
| Legal | .31** | — | |||||
| Health | .14** | .11* | — | ||||
| Economic | .09 | –.08 | .05 | — | |||
| Substance treatment | .14** | .01 | .14** | .15** | — | ||
| Counseling | .19** | .11* | .06 | .02 | .17** | — | |
| Religious–spiritual | .15** | .07 | .05 | .13* | .08 | .12** | — |
| Partner violence | |||||||
| Year psychological abuse | .22** | .13* | .00 | –.09 | .07 | .04 | .06 |
| Year physical abuse | .16** | .02 | .14** | –.04 | .07 | –.06 | .02 |
| Year sexual abuse | .14** | .15** | .12** | –.01 | .03 | .00 | .02 |
| Incident psychological abus | .15** | .18** | .22** | .03 | .14** | .03 | .06 |
| Incident physical abuse | .17** | .15** | .41** | .06 | .03 | –.02 | –.02 |
| Incident sexual abuse | .05 | .04 | .19** | .06 | .06 | .03 | .14** |
| Threat appraisals | .30** | .26** | .06 | –.04 | .02 | .02 | .05 |
| Biopsychosocial | |||||||
| Social positive | –.16** | –.03 | –.12* | –.08 | –.04 | .02 | –.07 |
| Social negative | .12* | .14** | .20** | .02 | .13* | –.01 | .10* |
| Depressive symptoms | .26** | .18** | .20** | –.05 | .08 | .08 | .11* |
| Physical health functioning | –.26** | –.22** | –.34** | –.05 | –.15** | –.10 | –.07 |
| Injuries at incident | .16** | .14** | .54** | .05 | .06 | .01 | –.01 |
| Alcohol abuse | –.04 | –.06 | .01 | –.04 | .21** | .01 | –.02 |
| Illegal drug use | –.05 | –.04 | .02 | .05 | .15** | .00 | .03 |
| Involved with perpetrator | .13* | –.08 | .10 | .07 | .08 | .13* | .02 |
| Demographics | |||||||
| Employment | –.04 | .04 | –.15** | –.38** | –.20** | –.03 | –.09 |
| Income | .07 | .17** | –.01 | –.36** | –.14** | .16** | .05 |
| Education | .03 | .14** | –.08 | –.27** | –.10* | .18** | .03 |
| Race/ethnicity | –.04 | –.11* | .10* | .14** | –.01 | –.14** | .04 |
| Age | .13* | .09 | .00 | –.08 | .05 | .16** | .11* |
p ≤ .05.
p ≤ .01.
p ≤ .001.
Although legal help-seeking was less extensively related to other forms of services help seeking, it shared a similar pattern with domestic violence help seeking with associations to violence exposure and biopsychosocial factors. However, unlike domestic violence help seeking, legal help seeking related to socioeconomic variables, with positive relationships to income and education and being European American. Health help seeking positively related to substance abuse, domestic violence, and legal services help seeking and to most forms of past-year and incident partner violence. Health help-seeking efforts were also significantly associated with employment and race or ethnicity. Although there was a positive relationship between being a woman of color and a report of health care help seeking, there was an inverse relationship between employment and health care help seeking.
Employment, income, and education inversely related to economic help seeking, and women of color were more likely than European American women to seek economic assistance. However, partner violence exposure, biopsychosocial, and demographic factors did not relate to economic help seeking. Substance abuse help seeking significantly and positively related to alcohol abuse, illegal drug use, negative social relationships, and impaired physical functioning. Substance abuse help seeking also positively related to incident psychological abuse and inversely related to the socioeconomic indicators of employment, income, and education. In addition to perpetrator relationship involvement at the time of the incident, counseling help-seeking efforts were most strongly associated with demographics, significant with all but employment. In a limited way, religious help-seeking efforts significantly related to sexual abuse at the incident, negative social relationships, and depression.
Forward stepwise regression analyses revealed that services help-seeking efforts were associated with distinct sets of needs and resources relevant to the nature of the respective services (Table 4). For example, women’s threat appraisals of their abusive partners significantly and positively accounted for variance in domestic violence, legal, and aggregated help-seeking efforts. Physical health functioning inversely accounted for domestic violence, legal, health care, and aggregated help-seeking efforts. Perpetrator involvement accounted for domestic violence and counseling help seeking and inversely related to legal help seeking. Injuries from the incident and alcohol abuse, respectively, accounted for health care and substance abuse services help seeking. Demographic factors also remained significant in the regression analyses, with employment and income inversely accounting for economic help seeking, education positively accounting for counseling help seeking, and employment inversely accounting for substance abuse help seeking.
Table 4.
Forward Stepwise Multivariate Regressions Relating Partner Violence, Biopsychosocial, and Demographic Factors to Battered Women's Services Help-Seeking Efforts(N = 382)
| Violence Help Seeking | SE | β Step 1 | β Step 2 | β Step 3 | |||
|---|---|---|---|---|---|---|---|
| Threat appraisals | .01 | .30*** | .25*** | .25*** | |||
| Physical health functioning | .01 | –.20*** | –.19*** | ||||
| Involved with perpetrator | .22 | .10* | |||||
| R2 | .09 | .12 | .14 | ||||
| F | 32.40*** | 23.66*** | 17.26*** | ||||
| Legal Help Seeking | SE | β Step 1 | β Step 2 | β Step 3 | β Step 4 | β Step 5 | β Step 6 |
| Threat appraisals | .01 | .26*** | .22*** | .20*** | .20*** | .17** | .21*** |
| Physical health functioning | .01 | –.17** | –.18*** | –.20*** | –.18*** | –.19*** | |
| Income | .07 | .17*** | .18*** | .17*** | .17*** | ||
| Involved with perpetrator | .27 | –.11* | –.14** | –.12* | |||
| Incident psychological abuse | .08 | .12* | .14** | ||||
| Physical abuse past year | .00 | –.13* | |||||
| R2 | .07 | .09 | .12 | .13 | .14 | .16 | |
| F | 23.65*** | 16.80*** | 15.10*** | 12.66*** | 11.10*** | 10.15*** | |
| Health Help Seeking | SE | β Step 1 | β Step 2 | β Step 3 | |||
| Injuries at incident | .03 | .54*** | .49*** | .48*** | |||
| Physical health functioning | .00 | –.14** | –.14** | ||||
| Race/ethnicity | .07 | .10* | |||||
| R2 | .30 | .31 | .32 | ||||
| F | 140.70*** | 76.09*** | 52.78*** | ||||
| Economic Help Seeking | SE | β Step 1 | β Step 2 | β Step 3 | |||
| Employment | .07 | –.38*** | –.30*** | –.29*** | |||
| Income | .03 | –.27*** | –.23*** | ||||
| Education | .04 | –.12* | |||||
| R2 | .15 | .21 | .23 | ||||
| F | 58.06*** | 45.19*** | 32.26*** | ||||
| Substance Help Seeking | SE | β Step 1 | β Step 2 | β Step 3 | |||
| Alcohol abuse | .05 | .21*** | .20*** | .20*** | |||
| Employment | .06 | –.19*** | –.18*** | ||||
| Incident psychological abuse | .03 | .12* | |||||
| R2 | .05 | .08 | .09 | ||||
| F | 15.79*** | 14.37*** | 11.33*** | ||||
| Counseling Help Seeking | SE | β Step 1 | β Step 2 | β Step 3 | β Step 4 | ||
| Education | .04 | .18*** | .16** | .17** | .15** | ||
| Age | .01 | .14** | .13** | .13* | |||
| Involved with perpetrator | .10 | .12* | .11* | ||||
| Race/ethnicity | .10 | –.11* | |||||
| R2 | .03 | .05 | .07 | .08 | |||
| F | 10.99*** | 9.10*** | 7.74*** | 6.90*** | |||
| Religious Help Seeking | SE | β Step 1 | β Step 2 | ||||
| Incident sexual abuse | .07 | .14* | .15** | ||||
| Age | .00 | .12* | |||||
| R2 | .02 | .03 | |||||
| F | 6.50* | 5.55** | |||||
| Aggregated Help Seeking | SE | β Step 1 | β Step 2 | β Step 3 | |||
| Physical health functioning | .02 | –.35*** | –.30*** | –.28*** | |||
| Threat appraisals | .02 | .21*** | .17*** | ||||
| Incident psychological abuse | .13 | .13** | |||||
| R2 | .12 | .16 | .18 | ||||
| F | 46.70*** | 32.32*** | 23.95*** | ||||
p ≤ .05.
p ≤ .01.
p ≤ .001.
Partner violence exposure remained significant in the multivariate analyses (Table 4). Incident psychological abuse accounted for legal, substance, and aggregated help seeking. There was also an inverse relationship between past-year physical abuse and legal help seeking. This inverse relationship is noteworthy because past-year physical abuse was not significantly related to legal help-seeking at the zero-order level (r2 = .02). This discrepancy suggests that past-year physical abuse has become a suppressor in the multivariate stepwise regression of the legal help-seeking index. An examination of the semipartial correlation of past-year physical abuse with legal help seeking, which was greater in magnitude and different in direction than the zero-order correlation, confirmed this assessment. Past-year physical abuse increased the prediction of legal help seeking for the overall model by controlling variance shared between past-year physical abuse and the other variables in the final regression step.
DISCUSSION
This study profiled the needs and resources of an urban sample of battered women related to services help seeking following an incident of partner violence that triggered either police involvement or a protection order. The findings suggest active efforts in help seeking by victims of partner abuse and attest to the multifaceted needs of battered women when they seek formal helping services. Somewhat reassuring was the finding that women who sought no services after the target incident differed from other battered women. Rather than being passive or overwhelmed, they appeared to be less acutely in need of formal interventions. They reported less extensive abuse in the preceding year and at the incident except on sexual abuse, lower appraisals of threat, less depression, less physical impairment, and fewer negative social relations.
The majority of the women in this sample did not seek specialized domestic violence services. More than one-third of the sample sought economic assistance and counseling services, and nearly two-thirds of the sample sought legal services. Although the percentages were smaller, women in this sample sought health, substance abuse treatment, and religious–spiritual services. In conjunction with earlier research that shows that battered women seek help for a range of life problems without calling attention to the violence in their lives (Henning & Klesges, 2002), these findings suggest the importance of practitioners across the broad spectrum of human services anticipating and screening for the needs of battered women who obtain their services. Although progress has been made in the extent to which various service settings screen for partner violence, this is not yet common practice (Levine & Greene, 2000). The purpose of this investigation was to assess the needs and resources battered women present with when seeking help. Thus, we cannot speak to the effectiveness of the services these participants sought. However, these data attest to the opportunity to intervene, especially with women with less severe histories for whom escalation may be prevented.
The investigation results have two overarching implications for social work assessment and intervention. First, battered women’s needs correspond with help seeking. Their needs, however, are not always presented with reference to the violence in their lives, and the majority do not reach domestic violence specialists. Battered women may seek economic and substance abuse services, for example, without ever reporting violence unless social workers ask about psychological, physical, and sexual violence. Social workers in all service settings should screen women for violence, otherwise critical intervention opportunities will be missed. Second, battered women who seek help for partner violence likely have a myriad of biopsychosocial difficulties resulting from the violence (Riger et al., 2002). The battered women contacting domestic violence services in this investigation were likely to be seriously depressed, have impaired physical functioning as well as injuries, and have few social relationships and activities that were positive and more that were negative. Practitioners who primarily provide domestic violence services should connect battered women with a range of supports and services to address co-occurring needs. Although violence cessation and safety planning are important intervention targets for these women, co-occurring biopsychosocial difficulties must be addressed concurrently.
Correlational relationships among the services categories improve our understanding of battered women‘s help seeking. Because the correlations vary across service categories, it is unlikely that an individual trait variable in proclivity to seek services can account for the patterns. Correlations of domestic violence services with other services are, comparatively, the strongest. Consistent with Brown’s (1997) analysis, one interpretation is that violence of greater duration and severity increases women’s contact with human services providers as part of a developmental process of seeing the battering behavior as a problem and women’s growing realization of the need for help. Women contacting domestic violence providers often have a range of needs (Sullivan, Basta, Tan, & Davidson, 1992). The correlations suggest that women are propelled to seek domestic violence services as partner violence escalates, and this contact facilitates connection with other service providers, likely through assessment and referral.
The array of biopsychosocial factors correlated with domestic violence, legal, and health care help-seeking efforts illustrates the complexity of adequately serving battered women. Our results indicate that women who seek these services have high levels of physical health, mental health, and social support needs. To facilitate safety for battered women, service providers must assess and address this range of biopsychosocial needs, because these problems may impede a woman’s ability to carry out a safety plan. A consistent finding in the correlational analyses is that negative social relationships positively relate to help-seeking efforts for many service categories. This adds a useful dimension to earlier findings that low social support is related to increased psychological distress (Thompson et al., 2000) and poor physical functioning in battered women (Nurius et al., 2003). The present findings suggest that negative social interactions may prompt a woman to seek formal services when she does not find adequate support in her social networks. These results argue for the utility of assessing the negative and positive aspects of social support in battered women.
The correlation results also have implications for social work with vulnerable and oppressed groups of women. Although women across socioeconomic lines sought help from domestic violence service providers, those seeking legal services were more likely to be white and have higher education and income levels. These findings indicate possible challenges for social workers helping women of color and lower incomes and education obtain legal services and reinforce the utility of providing legal services in domestic violence agencies and the importance of social work collaboration with legal services.
The multivariate analyses generally demonstrate that battered women’s needs drive their help-seeking efforts. For example, women seeking health care services experienced greater incident injury and difficulty with health functioning; similarly, substance abuse treatment help seeking is strongly related to alcohol abuse. Thus, human services providers are likely to encounter battered women with presenting problems consistent with their services and reinforce the importance of screening across human services settings. It is interesting that alcohol abuse relates to substance abuse help seeking, but that illegal drug use is not related at the multivariate level. This lack of significance may be due to the dichotomous measurement of this item. Thus, comprehensive drug use measurement is a recommendation for future research.
The multivariate analyses also provide information about where some of the most vulnerable groups of battered women appear for services. It is useful to know that women with fewer socioeconomic resources seek public assistance, women with lower employment levels seek substance services, women of color appear for medical treatment, older women seek help from counseling and religious services, and that women involved with the perpetrator seek help from domestic violence and counseling services. The results suggest avenues for targeting screening efforts for vulnerable populations such as socioeconomically disadvantaged, culturally diverse, and older women. Social workers in health care, public assistance, substance abuse treatment, religious, and counseling settings should have the skills to make referrals and provide resources to help women with safety. It is worrisome that women who have fewer socioeconomic resources or who are involved with their perpetrators may face barriers in obtaining access to other services (for example, legal services) that may facilitate safety. Targeted programming and outreach may be required to connect vulnerable women with needed services (Lee, Thompson, & Mechanic, 2002), such as placement of domestic violence specialists and legal advocates in neighborhood and public assistance centers and health care offices.
A striking finding is the extent to which threat appraisals and physical health functioning account for domestic violence, legal, health care and aggregated services help seeking. Recent research has made the case that, in addition to the behavioral features of abuse, partner violence must be understood in terms of vulnerability, loss of power and control, and entrapment (Smith et al., 1995). Our findings support this assertion. Specifically, threat appraisals strongly relate to domestic violence, legal, and aggregated help-seeking efforts. Related work has found that women’s fear and perception of threat relate to their completion of the restraining order process (Zoellner et al., 1998). Thus, a woman’s subjective experience of threat and danger from the perpetrator is a vital asset in linking her to supportive services, particularly those more directly able to augment her safety (Arias & Pape, 2001).
Physical health functioning is in part a measure of a woman’s ability to carry out daily life activities and tasks. It is noteworthy that greater difficulty in daily life functioning emerged as significant in several of the regression models. In combination with the importance of threat appraisals for help seeking, these results suggest that women seek help when they feel deeply threatened by their partners and when they have noticeable difficulties in their everyday lives because of the violence. When working with a battered woman, service providers should assess the extent to which a woman has difficulty with everyday physical functioning and feels threatened to determine how these difficulties impede her ability to carry out a safety plan. Feeling threatened by a partner and feeling unable to take care of life’s details may be motivators for help seeking. However, problems in these areas may also impair a woman’s ability to carry out safety and protective actions. Thus, coordination of legal and health care services is a critical safety action to help women with managing feelings of threat and enhancing physical functioning.
Not surprising, higher levels of partner violence remain significant in the multivariate analyses. Incident psychological abuse accounted for legal, substance, and aggregated help seeking. Similarly, incident sexual abuse accounted for religious help seeking. Consistent with other research (Sackett & Saunders, 2001), these findings show that the types of partner violence women experience affect their help-seeking efforts. The results suggest that women who appear for specific services like substance treatment and religious or spiritual services may bring with them high levels of psychological and sexual abuse. Optimally, social workers should include psychological and sexual abuse in their assessment and attend to the consequences of these types of abuse in interventions with battered women. Focus on physical abuse alone may not be sufficient in helping women with violence and safety.
Limitations
This research addressed a critical knowledge gap in understanding women’s help-seeking efforts in a relatively large sample of battered women across a range of human services. Still there are limitations to this investigation that help to highlight potential future research directions. Although the women who chose not to participate in this investigation were similar in many ways to the participants, they were less likely to have taken out protection orders and more likely to be living with the abuser. These findings suggest that nonparticipants may have been different from participants in ways that have implications for generalizability. There are many reasons why women do not seek help or are unwilling to participate in research. For example, perhaps the nonparticipants had greater positive informal supports than the participants. Alternatively, perhaps these women were more threatened and vulnerable than the participants and thus felt less safe to seek protection orders, end the abusive relationship, or participate in research. Future researchers may usefully follow up with nonparticipants to determine how these women differ from participants, if such efforts do not endanger women.
This sample is relatively diverse; however, the small sample sizes of specific groups of women of color and ethnicities may obscure important differences in help seeking. The women who came into this study were recruited on the basis of some kind of legal action. Although the woman did not have to make the call to the police herself, the results of this study may not generalize to battered women who belong to certain groups (for example, immigrant communities) or live in geographical areas (for example, rural) where the police may be less likely to be called because of partner violence. Oversampling of specific groups of women may be necessary to make meaningful comparisons among ethnic and racial groups. Cultural context, sociocultural factors, and geography may play a role in the kind and frequency of battered women’s help-seeking behaviors (Acevedo, 2000). Future research could examine how battered women from underrepresented communities define help seeking to incorporate actions that have not been considered here. In a related vein, the help-seeking indices were informed by earlier research in this area (Hutchison & Hirschel, 1998) and reflect formal services present in the community where the research was conducted. As a result, the help-seeking indices are not similar in their ranges given the number of service options available in specific categories; thus, certain types of help-seeking efforts are weighted more heavily. Future researchers may wish to investigate a wide array of social, health, and legal services, given that the types of services investigated here may or may not be present in all communities. Finally, future research should attempt to discern factors that explain help seeking in battered women. Although this investigation helped to determine the needs and resources that predict help seeking, the variance to be explained in the regressions suggests that women’s needs and resources alone cannot fully explain help seeking. Future research should consider factors beyond the battered woman’s experience and control, such as accessibility and availability of services, and investigate how a woman’s environmental and social context may shape her help-seeking efforts (Mitchell & Hodson, 1986).
It can be a challenge for social workers in a busy service environment to assess for partner violence when their primary task focuses on another area (McNutt, Carlson, Rose, & Robinson, 2002). Furthermore, the structural arrangements among different fields of practice or service systems can make it difficult for battered women to get their needs met and may also impede social workers from seeing “the whole woman” and her context (Gondolf, 1998). However, the findings from this investigation show that social workers across practice settings should assess for violence (Dienemann et al., 2000). At the community level, social workers should create links among service providers to promote successful outcomes for battered women no matter where they appear in a request for services. Although funding sources, agency missions, and target population parameters may impede cross-collaboration between social work providers at various agencies, a continuum of services may be needed by battered women to attain safety and to address other co-occurring difficulties. At a macro level, social workers can advocate for increased funding and collaboration across state and federal agencies to address the co-occurrence of partner violence with other problems such as poverty, mental illness, and substance abuse. Social programs and interventions are likely to be more effective when funding and policy reflect the multiplicity of problems with which battered women contend.
Acknowledgments
This investigation was supported by grant no. 1 R01 DA11151, Protection of Women: Health and Justice Outcomes, from the Centers for Disease Control and Prevention, National Institutes of Health, and National Institute of Justice, as part of the Interagency Consortium on Violence Against Women and Family Violence Research, and grant no. 5 T32 MH20010, Mental Health Prevention Research Training Program. We appreciate the contribution of Marsha Wolf in procurement of the original grant and the extensive research team in the conduct of this research. We also acknowledge Frederick P. Rivara and Maeda J. Galinsky for their contributions to this investigation and the comments of three anonymous reviewers.
Contributor Information
Rebecca J. Macy, Assistant professor, School of Social Work, University of North Carolina at Chapel Hill, 301 Pittsboro Street, CB No. 3550, Chapel Hill, NC 27599..
Paula S. Nurius, Professor and director, Doctoral Prevention Research Training Program, School of Social Work, University of Washington, Seattle..
Mary A. Kernic, Research assistant professor, Department of Epidemiology, School of Public Health and Community Medicine, University of Washington, Seattle..
Victoria L. Holt, Professor, Department of Epidemiology, School of Public Health and Community Medicine, University of Washington, Seattle..
REFERENCES
- Acevedo MJ. Battered immigrant Mexican women’s perspectives regarding abuse and help-seeking. Journal of Multicultural Social Work. 2000;8(3/4):243–282. [Google Scholar]
- Arias I, Pape KT. Psychological abuse: Implications for adjustment and commitment to leave violent partners. In: O’Leary KD, Maiuro RD, editors. Psychological abuse in violent domestic relations. New York: Springer; 2001. pp. 137–151. [Google Scholar]
- Barnett OW, Martinez TE, Keyson M. The relationship between violence, social support, and self-blame in battered women. Journal of Interpersonal Violence. 1996;11:221–233. [Google Scholar]
- Brown J. Working toward freedom from violence. Violence Against Women. 1997;3(1):5–26. doi: 10.1177/1077801297003001002. [DOI] [PubMed] [Google Scholar]
- Coker AL, Derrick C, Lumpkin JL, Aldrich TE, Oldendick R. Help-seeking for intimate partner violence and forced sex in South Carolina. American Journal of Preventive Medicine. 2000;19:316–320. doi: 10.1016/s0749-3797(00)00239-7. [DOI] [PubMed] [Google Scholar]
- Cooper P, Osborn M, Gath D, Feggetter G. Evaluation of modified self-report measure of social adjustment. British Journal of Psychiatry. 1982;141:68–75. doi: 10.1192/bjp.141.1.68. [DOI] [PubMed] [Google Scholar]
- Dienemann J, Boyle E, Baker D, Resnick W, Wiederhorn N, Campbell J. Intimate partner abuse among women diagnosed with depression. Issues in Mental Health Nursing. 2000;21:499–513. doi: 10.1080/01612840050044258. [DOI] [PubMed] [Google Scholar]
- El-Bassel N, Gilbert L, Schilling R, Wada T. Drug abuse and partner violence among women in methadone treatment. Journal of Family Violence. 2000;15:209–228. doi: 10.1007/s10896-008-9183-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Golding JM. Intimate partner violence as a risk factor for mental disorders: A meta-analysis. Journal of Family Violence. 1999;14:99–132. [Google Scholar]
- Gondolf EW. The victims of court-ordered batterers. Violence Against Women. 1998;4:659–678. [Google Scholar]
- Henning KR, Klesges LM. Utilization of counseling and supportive services by female victims of domestic abuse. Violence and Victims. 2002;17:623–636. doi: 10.1891/vivi.17.5.623.33714. [DOI] [PubMed] [Google Scholar]
- Hutchison IW, Hirschel JD. Abused women: Help-seeking strategies and police utilization. Violence Against Women. 1998;4:436–456. [Google Scholar]
- Lee RK, Thompson VLS, Mechanic MB. Intimate partner violence and women of color: A call for innovations. American Journal of Public Health. 2002;92:530–534. doi: 10.2105/ajph.92.4.530. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Levine SA, Greene JA. Linking substance abuse treatment and domestic violence services: A guide for administrators (DHHS Publication No. SMA00-3391) Rockville, MD: Substance Abuse and Mental Health Services Administration; 2000. [Google Scholar]
- Lincoln KD. Social support, negative social interactions, and psychological well-being. Social Service Review. 2000:231–252. doi: 10.1086/514478. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lloyd S. The effects of domestic violence on women’s employment. Law and Policy. 1997;19:139–167. [Google Scholar]
- McNutt LA, Carlson BE, Rose IM, Robinson DA. Partner violence intervention in the busy primary care environment. American Journal of Preventive Medicine. 2002;22(2):84–91. doi: 10.1016/s0749-3797(01)00407-x. [DOI] [PubMed] [Google Scholar]
- Mitchell RE, Hodson CA. Coping and social support among battered women: An ecological perspective. In: Hobfoll SE, editor. Stress, social support, and women. Washington, DC: Hemisphere; 1986. pp. 153–169. [Google Scholar]
- Nurius PS, Asplund LM. Community responses to partner abuse: Challenges and opportunities for social work and feminists. Social Policy and Social Work Practice. 1994;18:17–31. [Google Scholar]
- Nurius PS, Macy RJ, Bhuyan R, Holt VL, Kernic MA, Rivara FP. Contextualizing depression and physical functioning in battered women: Adding vulnerability and resources to the analysis. Journal of Interpersonal Violence. 2003;18:1411–1431. doi: 10.1177/0886260503258033. [DOI] [PubMed] [Google Scholar]
- Petersen R, Gazmarariam J, Andersen K. Partner violence: Implications for health and community settings. Women’s Health Issues. 2001;11:116–125. doi: 10.1016/s1049-3867(00)00093-1. [DOI] [PubMed] [Google Scholar]
- Plichta SB, Falik M. Prevalence of violence and its implications for women’s health. Women’s Health Issues. 2001;11:244–258. doi: 10.1016/s1049-3867(01)00085-8. [DOI] [PubMed] [Google Scholar]
- Riger S, Raja S, Camacho J. The radiating impact of intimate partner violence. Journal of Interpersonal Violence. 2002;17:184–205. [Google Scholar]
- Russell M, Martier SS, Sokol RJ, Mudar P, Bottoms S, Jacobson S, Jacobson J. Screening for pregnancy risk drinking. Alcoholism: Clinical and Experimental Research. 1994;18:1156–1161. doi: 10.1111/j.1530-0277.1994.tb00097.x. [DOI] [PubMed] [Google Scholar]
- Sackett LA, Saunders DG. The impact of different forms of psychological abuse on battered women. In: O’Leary KD, Maiuro RD, editors. Psychological abuse in violent domestic relationships. New York: Springer; 2001. pp. 197–211. [Google Scholar]
- Smith PH, Earp JAL, DeVellis R. Measuring battering: Development of the Women’s Experience with Battering (WEB) Scale. Women’s Health: Research on Gender, Behavior, and Policy. 1995;1:273–288. [PubMed] [Google Scholar]
- Straus MA, Hamby SL, Boney-McCoy S, Sugarman DB. The revised Conflict Tactics Scales (CTS2) development and preliminary psychometric data. Journal of Family Issues. 1996;17:283–316. [Google Scholar]
- Sullivan CM, Basta J, Tan C, Davidson WS., II After the crisis: A needs assessment of women leaving a domestic violence shelter. Violence and Victims. 1992;7:267–275. [PubMed] [Google Scholar]
- Thompson MP, Kaslow NJ, Kingree JB, Rashid A, Puett R, Jacobs D, Mathews A. Partner violence, social support, and distress among innercity African American women. American Journal of Community Psychology. 2000;28:127–143. doi: 10.1023/A:1005198514704. [DOI] [PubMed] [Google Scholar]
- Tjaden P, Thoennes N. Prevalence and consequences of male-to-female and female-to-male intimate partner violence as measured by the National Violence against Women Survey. Violence Against Women. 2000;6:142–161. [Google Scholar]
- Ware JEJ, Kosinski M, Keller SD. SF-12: How to score the SF-12 physical and mental health summary scales. Boston: New England Medical Center, Health Institute; 1995. [Google Scholar]
- Weissman MM, Paykel ES. The depressed woman: A study of social relationships. Chicago: University of Chicago Press; 1974. [Google Scholar]
- Wolf ME, Holt VL, Kernic MA, Rivara FP. Who gets protection orders for intimate partner violence. American Journal of Preventive Medicine. 2000;19:286–291. doi: 10.1016/s0749-3797(00)00235-x. [DOI] [PubMed] [Google Scholar]
- Wolf ME, Ly U, Hobart MA, Kernic MA. Barriers to seeking police help for intimate partner violence. Journal of Family Violence. 2003;18(2):121–129. [Google Scholar]
- Zoellner LA, Feeny NC, Avarez J, Watlington C, O’Neill ML, Zager R, Foa E. Factors associated with completion of the restraining order process in female victims of partner violence. Journal of Interpersonal Violence. 1998;15:1081–1099. [Google Scholar]