Social Work

J Fam Viol (2006) 21:271–279 DOI 10.1007/s10896-006-9021-7

O R I G I N A L A RT I C L E

Relationship Between Two Types of Help Seeking Behavior in Domestic Violence Victims Jean H. Hollenshead · Yong Dai · Mary Katherine Ragsdale · Erin Massey · Rachel Scott

Published online: 15 September 2006 C© Springer Science+Business Media, Inc. 2006

Abstract A coordinated community response system to the help-seeking behaviors of domestic violence victims is critical to minimize the impact of violent events and to ed- ucate the public, so that safe and effective conflict manage- ment skills may replace violent responses. The focus for the present study is to identify and analyze victims’ choices of law enforcement assistance in stopping the violence and/or aid through the services of the regional family violence center. Some victims select legal channels of support; oth- ers rely exclusively on social service support, and others seek assistance from both sources. Characteristics of victims whose cases followed two types of help-seeking behavior patterns—legal support or social support–were determined through archived data from both the records of a regional family violence center (FVC) (n = 258) and domestic vio- lence incident reports of a police department serving a city of approximately 200,000 citizens (n = 127). Findings indi- cate that African American victims more frequently sought protection through law enforcement sources than they did through family violence center services, and the reverse was true for Euro-Americans.

Keywords Help-seeking . Domestic violence .

Intervention

It has been noted that, throughout the world, the most salient risk factor for becoming a victim of violence is to be a woman (Walker, 1999). Domestic violence has become so prevalent that approximately 17% of the 1.4 million violence-related

J. H. Hollenshead (�) · Y. Dai · M. K. Ragsdale · E. Massey · R. Scott Department of Psychology, One University Place, Shreveport, Lousiana, USA e-mail: [email protected]

injuries seen in emergency departments are being inflicted by intimates (Rand, 1997). Women residing in Louisiana are more likely than women living elsewhere in the United States to be killed by a male, and too often the attacker is someone the victim knows. The incidence rate in Louisiana for female victims being murdered by males in single victim/single of- fender occurrences in 1996 was 2.93 per 100,000 women (Violence Policy Center, 1999). A similar study by the VPC (1999) showed that in 1997 Louisiana ranked first in the nation in such incidents, with an incidence rate of 3.94 per 100,000 women in the state. This is almost three times the na- tional average of 1.40 per 100,000 (VPC, 1999), suggesting an alarming increase in a single year.

Many other published studies document the magnitude and impact of domestic violence. Sullivan and Bybee (1999) have noted numerous studies that estimate that 21–34% of all women will be assaulted by a male partner at some point in their lives. Such violence in the home is also a pervasive and often unrecognized cause of mental illness, as well as phys- ical injuries or illnesses (Krishnan, Hilbert, VanLeeuwen, & Kolia, 1997). Injury resulting from a spouse or lover’s abuse is the single most common reason women enter hospital emergency departments (Mills, 1996).

Legal services

Traditionally, domestic violence has been viewed by the legal system as a problem that should remain within one’s home, rather than be resolved through the judicial system. Years of advocacy by women’s groups have prompted change in the legal system’s perception and handling of domestic violence incidents (Mills, 1998). Partially as result of this movement, in 1994, the U.S. Government enacted the Violence Against Women Act, which states that domestic violence is a human rights violation, which may be prosecuted under federal civil

Springer

272 J Fam Viol (2006) 21:271–279

rights laws. The act allows women to seek retribution for the loss of safety from violence in both the community and the home (Walker, 1999).

This act includes the Arrest Policies Program, which en- courages law enforcement agencies to adopt a mandatory or pro-arrest action in cases of domestic violence (National Institute of Justice, 1997). Mandatory arrest policies essen- tially move the discretion from law enforcement to the courts (Davis & Smith, 1995). The movement toward mandatory ar- rest, or pro-arrest, policies had its beginning in the 1984 Min- neapolis Domestic Violence Experiment, which found that arresting domestic violence perpetrators reduced the risk of future assaults by one half over a six-month follow-up period (Schmidt & Sherman, 1993).

In conjunction with mandatory arrest policies, some legis- latures have enacted no-drop prosecution policies. The argu- ment for mandatory prosecution is partially based upon the assumption that an abuser may manipulate his victim into dropping charges, but future violence may be deterred if he knows that police and prosecutors will pursue the case re- gardless of the victim’s wishes (Wattendorf, 1996). The move toward relieving the victim of the responsibility of prosecu- tion has resulted in a debate concerning the efficacy of such a process, as well as an effect on the safety of the victim (Davis & Smith, 1995). On the local level, Susan Reno, of a metroplititan city police department (personal communica- tion, August 14, 1999), has reported that since the installation of a local no-drop prosecution policy in 1997, the number of reported incidents has dropped by one half.

Arrest and prosecution policies, enacted to reduce new cases of domestic violence and recidivism of previous of- fenders, are still being studied and their efficacy debated. Although the focus in recent years has concentrated largely on law enforcement and judicial system interventions, other effective alternatives exist in the form of community-based intervention. These community-based interventions allow battered women access to a variety of programs, including legal advocacy, counseling, court orders, residential shelters, and transitional housing.

Social support services

Social support and community resources have often been stressed as effective means of enabling women to escape an abusive relationship. Factors that increase a woman’s risk of abuse by partners and ex-partners include social isola- tion of the woman by her abuser, as well as an ineffec- tive community response to domestic violence (Sullivan & Bybee, 1999). A wide variety of formalized services and informal networks exist. These services can include com- munity services or self-help services aimed at empowering the battered woman. Early research by Bowker (1988) indi- cated that, regardless of many avenues of assistance available

to battered women, 35–45% of abused women never seek help outside of contact with family and friends (Hotaling, Finkelhor, Kirkpatrick, & Straus, 1988).

Gordon (1996) has reported that the most frequently con- tacted formal sources of assistance are the police, social service agencies, crisis counselors, physicians, psycholo- gists, counselors, lawyers, women’s groups, and the clergy. In Bowker’s (1988) study of 1,000 abused women, 39% of the women found the police to be very helpful in reduc- ing or stopping their abuse; however, 19% reported an in- crease in violence due to police intervention. Fifty percent of the abused women reported lawyers to be more helpful than psychologists, psychiatrists or the clergy in ending the abuse (Hotaling, Finkelhor, Kirkpatrick, & Straus, 1988). Several studies have indicated that although support groups and battered women’s shelters are some of the most spe- cialized services offered to domestic violence victims, they are often the least contacted (Gordon, 1996). Earlier stud- ies have suggested that this may be attributable to a lack of availability of such services in certain areas, or to the stigma women experience when revealing their abusive his- tories (Gordon, 1996). Despite the fact that support groups and battered women’s shelters are often the least contacted of battered women’s services, Bowker found they are often rated by abused women as the most helpful and effective means of coping with abuse (Gondolf & Fisher, 1988). Bat- tered women’s shelters, in particular, provide a sanctuary for women, as well as a variety of services such as legal advocacy, counseling, women’s groups, and transitional liv- ing arrangements. According to one study conducted in a battered women’s shelter, more than half of all subjects be- came independent after their stay at the shelter (Krishnan et al., 1997).

Legal advocates work with battered women in an effort to assist them in gaining independence from their abusers. In a study of battered women and police response by Jaffe, Hastings, Reitzel, & Austin (1993), it was found that the victims wished for more information on judicial processes and available community services. Psychoeducational sup- port and networking are critical to the battered woman’s autonomy. The efficacy of this support is noted in a study by Sullivan and Bybee (1999), which found that women who received free advocacy services reported less physi- cal violence, increased quality of life, lessened depressive symptoms, and increased ability to obtain resources when compared to women who had not received such services.

Legal advocacy has been an influential force for abused women, in that it provides them a resource to assist in ob- taining restraining orders. Although restraining orders are not the only available avenues for battered women, they are often utilized as protective barriers to eliminate fur- ther violence or contact between a victim and an abuser. Research has produced equivocal results in regard to the

Springer

J Fam Viol (2006) 21:271–279 273

utility of restraining orders. On one hand, they may be viewed as a legal tool available to victims of violence; on the other hand, they may be limited in breadth, thereby not truly pre- cluding contact (Gondolf, McWilliams, Hart, & Stuehling, 1994) or they may be poorly enforced (Holmes, 1993).

Simon (1995) indicates that restraining orders (including protective orders, a subset of a restraining order) often give victims a false sense of security. This vulnerability is at- tributed to the legal system’s frequent failure to effectively enforce restraining orders, as well as to the inherently in- creased risk of harm to women who attempt to separate from their abusers. Unfortunately, restraining orders are often in- effective when enacted against abusers with prior histories of arrest (Keilitz, 1994; Keilitz, Davis, Efkeman, Flango, & Hannaford, 1998). Another weakening aspect of restraining orders lies in the ability of abusers to intimidate their victims, thereby precluding them from fully enacting the order (The Urban Institute, 1995–1996).

Despite their potential shortcomings, Simon (1995) states that restraining orders are often effective in providing imme- diate relief to women and states that their usefulness depends on a number of factors: “. . .specificity of relief ordered; en- forcement practices of the police and courts; the severity of violence in the year prior to the order; and, the level of re- sistance by the man during the hearing” (p. 53). Therefore, the true strength in restraining orders appears to be a tem- porary relief from violence during which period the battered woman may take more permanent, effective steps toward a life independent of her abuser (Simon, 1995).

Interventions not only provide information to battered women in legal matters; they also enable women to seek assistance in the form of counseling, or self-help groups. Battered women’s groups are helpful in raising self-esteem and allowing women to become more independent of their abusers. These groups have been shown to be effective means of stabilizing a battered woman’s life while she goes through the process of separating from her abuser (Gordon, 1996). Tutty and colleagues studied 12 battered women’s support groups and found that participation in a support group spurred a number of positive outcomes, such as in- creased self-esteem, a reduction in perceived stress, depres- sion, and anxiety, and an improved perception of marriage and the family (Tutty, Bidgood, & Rothery, 1993).

Despite the many obstacles that battered women face, such as social isolation, lack of community support, and financial dependence, they are often found to be active help-seekers. A comprehensive study conducted by Gondolf in 1988 found that the women in battered shelters had made an average of six previous attempts to obtain relief from their abuse (Gondolf & Fisher, 1988). Sullivan (1991) stated that three factors seem to influence the decision of a battered woman to seek help: the severity of abuse, the number of resources available to the woman, and the woman’s own perceptions of

the effectiveness of these efforts (Sullivan & Bybee, 1999). Unfortunately, the time when women are most likely to be ac- tive help-seekers is early in the abusive relationship. A study by Reidy and Von Korff (1991) found that an increased sever- ity of abuse was often related to longer delays in seeking as- sistance. The authors suggest that social isolation could play a possible role as abuse escalates. Severe abuse compounds their feelings of isolation from available resources, as well as increased feelings of helplessness, which, in turn, reinforce the reluctance to seek help to end an abusive relationship (Reidy & Von Korff, 1991).

Little research has been conducted concerning help- seeking efforts of domestic violence victims from minority populations (Krishnan et al., 1997). Although domestic vio- lence does not limit itself to any particular racial group, age, or socioeconomic status, there are differences among mi- nority populations in the experience of domestic violence. Victims come from different backgrounds and may be un- der additional strain due to cultural practices and influences (Krishnan et al., 1997). These researchers also found that more Anglo women reported abuse incidents to law enforce- ment or sought medical attention than did Hispanic victims. Similarly, a possible cultural factor was demonstrated in the work of Neighbors and Jackson (1984), in which African Americans were found to have not sought assistance from mental health providers, but instead relied upon informal support networks or sought no assistance at all. These find- ings may suggest a cultural difference in perceived accessi- bility and/or relevance of community sources of support.

Ethnicity and choices of interventions

In Gondolf and Fisher’s study (1988), no differences in fre- quency of domestic violence were found between racial pop- ulations, although a significant finding was that Hispanic women often experienced a longer duration of abuse than African American or Euro-American women before seek- ing help. The investigators also noted that Euro-American battered women were more likely to report abuse to law enforcement or seek medical attention than were Hispanic women. In this particular study, Hispanic women tended to have fewer social and economic avenues to gain indepen- dence from their abusers.

A Canadian study analyzing ethnicity and help-seeking patterns in Native Indians with severe mental illness yielded results indicating that available social and community ser- vices were seriously underutilized by those most in need (Lin, Tardiff, Donetz, & Goresky, 1978). In addition, this study noted that this group was characterized by low so- cioeconomic status and minimal social support networks. Assistance received by the Native Indians in this study was frequently an outcome of a conflict with the law, rather than

Springer

274 J Fam Viol (2006) 21:271–279

the result of voluntary help-seeking. A few investigators (Poelzer & Poelzer, 1986) have documented a lack of sys- tematic, descriptive research focused on the Native American population in the context of an alarming increase in domestic violence within this group.

The majority of literature focusing on ethnicity and help-seeking behaviors has involved African Americans. Sussman, Robins, and Earls (1987) analyzed treatment seek- ing by clinically depressed African Americans. The authors found that African Americans were less likely than Euro- Americans to seek help for problems that were not considered severe. They also found a relationship between the break- down of familial support networks and African Americans’ decision to seek treatment. More often than not, a breakdown in familial networks was positively correlated with the deci- sion to seek treatment. Browman (1987) found that African Americans were more likely than Euro-Americans to uti- lize mental health services. The African Americans in this study overwhelmingly sought assistance from mental health services or other professional services, excluding medical as- sistance or clergy. The African Americans in this study were more likely to seek assistance at the lowest level of distress; however, when the level of distress was approximately equal, there was no difference in help-seeking between African Americans and Euro-Americans.

In a literature review of battered African American women, Coley and Beckett (1988) indicated that when other factors than race were controlled for, there were no major differences in the incidence of battering between African Americans and Euro-Americans. In other words, race alone was not empirically supported as a primary factor in distin- guishing violent from nonviolent couples. However, different findings emerged when Cazenave and Straus (1979) analyzed their data while controlling for factors such as gender. For example, African American women reported less violence than did Euro-American women.

Despite these equivocal data, recent, sobering data from FBI files, as reported by the VPC (1999), reflect a rate of single female victim/single male offender homicide in 1997 that is almost four times greater for African American women (3.88 per 100,000) than for Euro-American women (1.01 per 100,000) nationwide. In Louisiana, out of 89 female murder victims, 61 were African American, 27 were Euro-American, and 1 was American Indian or Alaskan Native.

The present study

The purpose of the present study is to assess the help-seeking behavior of domestic violence victims through an analysis of relationships among a variety of personal /situational vari- ables and two types of help-seeking behavior. This was ac- complished by a review of archival records of a regional

FVC and a metropolitan police department. This study fo- cuses on the selection of law enforcement intervention and/or community-based social service and advocacy intervention services by abused women.

This research seeks to identify factors that may play a role in the choice of help-seeking behaviors in the domes- tic violence context. More specifically, it is the intent of the present study to investigate the possible relationship be- tween an individual’s race or age and one’s help-seeking behavior. The decision to involve either law enforcement or community-based advocacy services is a crucial one in a bat- tered woman’s life, and research concerning this particular area is sparse, yet sorely needed, in order to more effec- tively serve battered women. Specific research questions posed are:

1. Is there a disproportionate utilization rate for two types of help-seeking behavior (law enforcement and social services) as a function of age, gender, or ethnic group membership? Research has suggested an underutiliza- tion of specialized services designed to assist the do- mestic violence victim (Gordon, 1996). Related studies have indicated widely varying preferences in sources of assistance—police, shelters, mental health and clerical professionals, and support groups (Gondolf & Fisher, 1988). However, few researchers have attempted to iden- tify basic demographic groupings that may differen- tiate one preference from another. Therefore, this lo- cal population was selected to initiate an investigation into such variables as age, gender, and ethnic group membership.

2. Is there a relationship between the identification of the complainant (victim, abuser, or third party) as identified in police incident reports and factors such as the type of abuse alleged, rate of arrest, type of arrest (single/dual), or age of the identified victim? Recent studies have indicated equivocal findings in regard to the efficacy of mandatory or pro-arrest policies encouraged by the 1994 Violence against Women Act (Schmidt & Sherman, 1993). Be- cause the present database includes a considerable num- ber of cases (n = 37) in which the complainant was a third party, the question arises as to whether or not the identity (role) of the complainant (third party, victim, or abuser) is related to the subsequent handling of the case by the police, i.e., no arrest, single arrest, or dual arrest.

3. Is there a relationship between the type of social service used through a domestic violence shelter and the age or ethnic group membership of the abuse victim? Findings suggesting an overall underutilization of mental health services by minority group members (Corey, 2000) en- gender the question of which types of available services may be most acceptable to which groups of abuse vic- tims. Local services range from those focusing primarily

Springer

J Fam Viol (2006) 21:271–279 275

on physical safety (shelter, restraining order assistance) to those covering broader issues, such as empowerment and self-image reconstruction (outpatient counseling, support groups). Are these different types of services selected as more or less relevant and/or acceptable by different groups of victims, according to age or ethnicity?

Method

Program descriptions and definitions

In order to examine the utilization of the two types of help-seeking behaviors of domestic violence victims, it was necessary to locate a data set from primary intervention agencies, representing the two types of help-seeking under study. Therefore, archival data from a regional FVC and a metropolitan police department served as data sources. The FVC provides five types of services. These services are out- patient counseling, shelter services, transitional housing, and two types of legal advocacy services, one that includes as- sistance in the process of obtaining restraining orders, and another that involves courtroom support when victims must confront their abusers in order to effect prosecution. Outpa- tients are defined as clients who receive individual or group counseling services. Residents include clients of the FVC seeking temporary shelter. Transitional housing clients are those individuals who have exceeded the maximum 30-day stay in residential housing, and require continuing residential services.

The metropolitan police department is the law enforce- ment agency responsible for responding to domestic violence reports within the city. Therefore, incidents reported to the department all occurred within the city limits, a somewhat more restricted geographical area than that served by the FVC.

For the purpose of this study the various types of abuse are operationally defined as:

1. Physical abuse—slapping, shaking, punching, beating, biting, pushing, pulling hair, pinching, kicking, stab- bing, tickling excessively, threatening, abandonment in dangerous situations, property/pet destruction, wielding weapons, use of weapons, denials of medical care, driv- ing recklessly, unwanted sexual acts, forced sexual acts, and/or forced prostitution.

2. Nonphysical abuse—emotional, verbal or mental abuse including accusation of infidelity, humiliation, name call- ing, isolation, locking out of the home, manipulation, con- tinuous criticism, restricting from work, and/or control- ling money.

3. Weapons—use of weapons is included within the physical abuse category, but data was gathered to identify subsets of physical abuse—with weapons or without weapons.

Participants

Participants for the present study included 131 individuals who filed formal complaints of domestic violence with a law enforcement intervention unit (LEI), and 259 clients of a regional FVC. Included were four participants who sought services through both the LEI and the FVC. A domestic violence incident is defined by the LEI as “physical abuse between cohabitants” (K. L. Skeesick, personal communi- cation, October 29, 1999), and the FVC defines domestic violence as “a pattern of coercive control exercised by one intimate partner over the other.” These definitions served as the criteria for inclusion of a participant. All individuals who entered the two protective systems and for whom adequate data were available between August 1, 1998, and February 28, 1999, were included. The services provided by the FVC to domestic violence victims in the present study include: outpatient counseling (n = 100), 30-day shelter residence (n = 56), legal advocacy – restraining orders (n = 107), le- gal advocacy – court representation (n = 5), and transitional housing (n = 10). The total number of client services for the participant pool was 278 services provided to 259 clients for the 7-month period. These records include 15 individuals who received 2 types of services, and two individuals that re- ceived three separate services. LEI incident reports include a category entitled “assaults,” and domestic violence inci- dents are included in this assault category as those involving reports of physical abuse by cohabitants.

Demographics

The sample was grouped according to age at the time of the incident report and/or FVC service, gender, and ethnic group membership. Age groupings were identified as follows: 22 and below (n = 57); 23 through 29 (n = 96); 30 through 39 (n = 149); 40 and over (n = 79); and Unknown (n = 5). The group included 376 females (97.4%) and 10 males (2.6%).

According to ethnic grouping, included were African Americans (n = 191; 49.5%), Euro-Americans (n = 179; 46.4%), Asian Americans (n = 5; 1.3%), Hispanics (n = 3; 0.8%), Other (n = 1; 0.3%), and Unknown (n = 7; 1.8%). There were no Native Americans in the participant pool. The ethnic group membership of the participant pool was compared to data generated by the U.S. Bureau of Census (1998), which estimates the ethnic composition of the city as follows: African Americans = 47.2%; Euro-Americans = 51.6%; Asian Americans = .7%; Hispanics = 1.4%; and Native Americans = .2%.

Procedures

For this ex post facto study, data were collected from FVC records of services available for the period August 1, 1998,

Springer

276 J Fam Viol (2006) 21:271–279

Table 1 Demographic comparisons between the Metropolitan City and the FVC Region 7 based on 1998 Census Bureau

Population size Population size in Ethnicity in the citya % FVC Region 7 % Sample size %

African American 89,700 47.2 169,501 33.8 191 49.5 Euro-American 98,000 51.6 276,355 55.1 179 46.4 Asian American 1,300 0.7 3,030 0.6 5 1.3 Hispanic 2,700 1.4 6,845 1.36 3 0.8 Native American 400 0.2 1,128 0.23 0 0.0

aDenotes numbers rounded to the nearest hundredth.

through February 28, 1999. All clients of this facility were included in the study, provided that sufficient data had been collected via the FVC Non-Residential Intake Form, FVC Resident Intake Form, FVC Domestic Abuse Relief As- sistance Intake Form, FVC Legal Advocacy Form, and the FVC Thresholds – Application for Residency. A FVC vic- tim advocate, who serves as a liaison between the FVC and the LEI, provided incident report data for the present study. Records from both sources were sanitized by the use of coded identifiers to protect the confidentiality of participants.

All data were grouped and coded on a data collection sheet designed specifically for the project. All information from the data collection sheets was subsequently posted to a data table, which was then sorted by codes to match participants’ records for multiple services, multiple incident reports, and cases of incidents and services. Subsequent to this matching process, identifying codes were removed from the database in order to protect the confidentiality of the participants. Data used for analysis included demographic information, type(s) of abuse, relationship of abuser and victim, abuse history, incident dates and circumstances (arrest conditions, identification of complainant, and removal of victim/abuser), and FVC services provided (number, types, and dates of services).

Results

Coded and grouped data were analyzed using the Pearson chi-square test for independence to test for associations be- tween help-seeking behavior of domestic violence victims and age, ethnic group membership, gender, or role of the complainant in police-reported incidents. An α level of .05 was used for all statistical tests.

Four of the age groups were included in the chi-square test with the omission of the unknown age group. There was no statistically significant association between age groups and help-seeking behavior of domestic violence victims. Be- cause of the limited sample sizes for other ethnic groups, only African American and Euro-American were used in the Pearson chi-square test for independence. A statistically sig- nificant association was demonstrated between ethnic group membership (African American or Euro-American) and the

type of intervention (LEI or FVC) selected (χ 2(1) = 33.457, p < .001). Examination of adjusted residuals indicated that African Americans sought assistance through law enforce- ment channels significantly more often than they turned to social services available through a family violence center. The reverse was true for Euro-Americans, with a demon- strated preference for family violence social services over those of law enforcement. Results based on the odds ratio (θ = 3.83) indicated that for African American victims, the odds of seeking law enforcement intervention were about four times the odds of seeking law enforcement intervention by Euro-American victims (Tables 1 and 2).

A related finding was a statistically significant association between ethnic group membership and the type of service used through FVC when FVC was selected (χ 2(3) = 11.924, p < .01) (see Table 3). Adjusted residuals indicate that among those African Americans who did select FVC services as their source of assistance, the type of service sought from that source was more frequently restraining orders through the court, with the assistance of the FVC advocate, than it was outpatient counseling. The reverse was true for Euro- Americans, who more frequently selected outpatient coun- seling than they did restraining orders. The odds of seek- ing restraining orders by African Americans were two to three times the odds of seeking restraining orders by Euro- Americans (θ = 2.58).

Table 2 Cross-tabulation of type of intervention utilized with ethnic group of victim

Law Family violence enforcement center Total

Ethnicity African American

Observed count 91 97 188 Expected count 64.7 123.3 188.0 Adjusted residual 5.8 −5.8

Euro-American Observed count 35 143 178 Expected count 61.3 116.7 178.0 Adjusted residual −5.8 5.8

Total Observed count 126 240 366 Expected count 126.0 240.0 366.0

Note. χ 2(1) = 33.457, p < .001.

Springer

J Fam Viol (2006) 21:271–279 277

Table 3 Cross-tabulation of type of service and ethnic group of victim

African American Euro-American Total

Service type Outpatient

Observed count 24 66 90 Expected count 36.6 53.4 90.0 Adjusted residual −3.4 3.4

Resident Observed count 25 24 49 Expected count 19.9 29.1 49.0 Adjusted residual 1.7 −1.7

Court orders Observed count 47 50 97 Expected count 39.5 57.5 97.0 Adjusted residual 2.0 −2.0

Total Observed count 96 140 236 Expected count 96.0 140.0 236.0

Note. χ 2(2) = 11.924, p ≤ .01.

When the identity (role) of the complainant (e.g., third party observer)was considered in relation to the tendency of the LEI to result in an arrest, findings indicated that there was a greater frequency of arrest when the complainant was a third party than when he/she was the identified victim (χ 2(1) = 8.28, p < .01) (see Table 4). The odds of arrest of the abuser as a result of the third party intervention were seven times the odds of arrest of the abuser as a result of the victim’s report (θ = 7.02).

Similarly, there was also a greater frequency of a dual arrest when the complainant was a third party, rather than the identified victim (χ 2(2) = 20.104, p < .001) (see Table 5). The odds of a dual arrest as a result of a third party complaint were five times the odds of a dual arrest when the complaint was made by the identified victim (θ = 5.26).

When gender was examined in relation to intervention type (i.e., LEI vs. FVC), there is one cell in the contingency

Table 4 Cross-tabulation of role of complainant with arrest status

Yes No Total

Complainant Third party

Observed count 31 2 33 Expected count 24.9 8.1 33.0 Adjusted residual 2.9 −2.9

Victim Observed count 64 29 93 Expected count 70.1 22.9 93.0 Adjusted residual −2.9 2.9

Total Observed count 95 31 126 Expected count 95.0 31.0 126.0

Note. χ 2(1) = 8.287, p < .01.

Table 5 Cross-tabulation of role of complainant with type of arrest

Dual arrest

Abuser arrest

No arrest Total

Complainant Third party

Observed count 13 17 2 32 Expected count 5.5 18.7 7.8 32.0 Adjusted residual 4.1 −.7 −2.8

Victim Observed count 8 55 28 91 Expected count 15.5 53.3 22.2 91.0 Adjusted residual −4.1 .7 2.8

Total Observed count 21 72 30 123 Expected count 21.0 72.0 30.0 123

Note. χ 2(2) = 20.104, p < .001.

table that expected counts less than 5. Thus, Fisher’s exact test was performed to examine the association. A statistically significant relationship was noted between gender and inter- vention type (p < .0296). Examination of the data indicated that while all the male victims (n = 10) sought help through LEI, fewer females than expected utilized LEI assistance, but instead turned to FVC.

Tests for association yielded no significant results when age was tested in relation to type of service sought, identity of the complainant, or service type used within the FVC.

Discussion

The present study was designed to assess the choices in help-seeking behaviors in domestic violence victims, as a function of age, gender, ethnic group membership, or com- plainant role in law enforcement reported incidents. Overall, findings were consistent with the published literature indi- cating a strong underutilization of social services by minor- ity groups. Results indicated that African Americans in the present study, when given an option of an active law enforce- ment assistance program and a full range of family violence center services, more frequently selected the law enforce- ment option. Even when they did elect to use the family violence center services, the selected service was more fre- quently advocacy support to obtain restraining orders than it was counseling. The reverse was true for the Euro-American group, who more often used outpatient counseling services as a source of support than might have been expected by chance, and less frequently used advocacy to obtain restraining or- ders. These findings may reflect a greater familiarity with law enforcement protective services due to a greater visibility of law enforcement personnel in the African American envi- ronment in the local area than in that of the Euro-American

Springer

278 J Fam Viol (2006) 21:271–279

environment, and, thus, a greater comfort level in seeking their assistance.

A related finding pertains to the source of the complaint to law enforcement when a domestic violence incident is re- ported. Within the sample, there were 33 instances of a third party complaint to law enforcement, while the identified vic- tim filed the complaint in 93 cases. Data analysis showed that in cases where the complainant was a third party (i.e., police officer or other observer), there was a greater pro- portion of arrest as a result of LEI (93.9%) than when the victim reported the incident (68.8%). Interpretation of this finding could vary widely, as the circumstances of the third party’s involvement were undetermined. For example, po- lice officers may have a greater tendency to make an arrest if they witness the violence than when an unobserved incident is called in. Alternatively, it may be possible that an unin- volved witness to an abusive incident may be more prone to seeking the help of the police when the violent acts are more severe and place the victim at greater risk, than when the violence is less severe. Victims may be more willing to contact police early in an abusive encounter as a preventive, self-protective measure against possible further harm—an intervention less likely to result in arrest. Such interpreta- tions need to be further investigated in future studies.

Although the number of males in the sample was small (n = 10), Fisher’s exact test indicated significant findings, with male victims leaning toward LEI assistance (100%) as opposed to FVC. This finding, however, is probably of little practical significance, as the local family violence center is housed in a women’s service facility, and is widely perceived to be a service for battered women. However, alternate hy- potheses may revolve around the stereotypic hesitancy in men to seek protection from women, as this proactive stance may be perceived as masculine weakness. Again, the utiliza- tion of available social support services by male domestic violence victims needs to be explored fully, as there is a lack of studies pertaining to this group in the literature.

Limitations of the present study relate to such factors as sample size and the limited geographic regions included. Several groupings were too small to be included in the statis- tical analysis, such as the Asian American group, an identi- fied group of individuals seeking both LEI and FVC services, and the identified abuser as complainant group. The catch- ments areas included in the study are a variable to be con- sidered. The local FVC covers a geographical region, which was largely rural. The local police department, on the other hand, responds only to calls from the city, a largely urban area. Future investigations in this area may need to include sheriff’s departments from these rural areas, and other city police units within the region in order to balance geographic and urban/rural influences on the findings.

An additional limitation of the study is the inclusion of incident reports resulting from complaints by a third party.

The study was designed to assess help-seeking behaviors in victims, and it is unknown to what extent victims in the third party reported incidents even concurred with contacting law enforcement, much less voluntarily selected this resource as their choice of an avenue for protection. Therefore, in order to determine the actual level of victim involvement in the choice, further review of available data would need to be conducted to analyze more specifically the role of the third party complainant.

It should be noted that one of the original objectives of this study was to investigate the relationship of forms of arrest in domestic violence incidents (dual versus single arrest) to a victim’s willingness to seek further protective services, such as shelter services or counseling designed to empower vic- tims in regard to the continuation of an abusive relationship. However, this investigation was found to be impractical, with its present design, as only four of the 386 cases studied in- volved individuals who subsequently sought FVC services of any kind following LEI.

Questions are raised, therefore, regarding the reasons for such a lack of involvement in both types of interventions, in view of the considerable research indicating that a coordi- nated community response system in the area of domestic violence provides the greatest resiliency against future harm.

In conclusion, if research indicating that a coordinated community response system involving both social services and law enforcement has been demonstrated to be effective in empowering battered women, and that less than half of all battered women currently seek formal community sup- port services, then research must turn to inquiry into current sociocultural factors and norms that are resulting in a wide discrepancy between the existence of known beneficial re- sources and use by those most in need of those services. Similarly, if there has been a severe underutilization of so- cial support services by minority groups, particularly African Americans, then investigators must seek to better understand the dynamics that are preventing this group from accessing such services. Is there a perceived inaccessibility of com- munity supports among high-need groups due to lack of awareness of their existence, or insufficient proximity in a crisis situation? Are there beliefs about institutional assis- tance that preclude the minority battered woman from al- lowing helpers into her private and frightening world? These and related questions must be asked and answered in order to stem the loss of dignity and human fulfillment that results from domestic violence.

References

Bowker, L. (1988). The effect of methodology on subjective estimates of the differential effectiveness of personal strategies and help sources used by battered women. In G. Hotaling, D. Finkelhor, J. Kirkpatrick, & M. Straus (Eds.), Coping with family violence:

Springer

J Fam Viol (2006) 21:271–279 279

Research and policy perspectives (pp. 80–92). Newbury Park, CA: Sage .

Browman, C. (1987). Race differences in professional help-seeking. American Journal of Community Psychology, 15(4), 473–489.

Cazenave, N., & Straus, M. (1979). Race, class, network embeddedness and family violence: A search for potent support systems. Journal of Comparative Family Studies, 10(3), 281–300.

Coley, S., & Beckett, J. (1988). Black battered women: A review of empirical literature. Journal of Counseling and Development, 66, 266–270.

Corey, G. (2000). Theory and practice of group counseling (5th ed.). Belmont, CA: Brooks/Cole.

Davis, R., & Smith, B. (1995). Domestic violence reforms: Empty promises or fulfilled expectations? Crime and Delinquency, 41(4), 541–553.

Gondolf, E., & Fisher, E. (1988). Battered women as survivors: An alternative to treating learned helplessness. Lanham, MD: Lex- ington Books.

Gondolf, E. W., McWilliams, J., Hart, B., & Stuehling, J. (1994). Court responses to petitions for civil protection orders. Journal of Inter- personal Violence, 9, 503–517.

Gordon, J. (1996). Community services for abused women: A review of perceived usefulness and efficacy. Journal of Family Violence, 11(4), 315–329.

Holmes, W. M. (1993). Police arrests for domestic violence. American Journal of Police, 12, 101–125.

Hotaling, G., Finkelhor, D., Kirkpatrick, J., & Straus, M. (Eds.). (1988). Coping with family violence: Research and policy perspectives. Newbury Park, CA: Sage.

Jaffe, P., Hastings, E., Reitzel, D., & Austin, G. (1993). The impact of police laying charges. In N. Hilton (Ed.), Legal responses to wife assault (pp. 62–95). Newbury Park, CA: Sage .

Keilitz, S. (1994). Civil protection orders: A viable justice system tool for deterring domestic violence. Violence and Victims, 9(1), 79– 84.

Keilitz, S., Davis, C., Efkeman, H., Flango, C., & Hannaford, P. (1998). Civil protection orders: Victims’ views on effective- ness. (NIJ Grant No. 93-IJ-CX-0035). National Institute of Jus- tice. Retrieved August 2, 1999 from the World Wide Web: http://www.ncjrs.org/txtfiles/fs000191.txt.

Krishnan, S., Hilbert, J., VanLeeuwen, D., & Kolia, R. (1997). Docu- menting domestic violence among ethnically diverse populations: Results from a preliminary study. Family and Community Health, 20(3), 32–49.

Lin, T., Tardiff, K., Donetz, G., & Goresky, W. (1978). Ethnicity and patterns of help-seeking. Culture, Medicine, and Psychiatry, 2, 3–13.

Mills, L. (1996). Empowering battered women transitionally: The case for postmodern interventions. Social Work, 41(3), 261–267.

Mills, L. (1998). Mandatory arrest and prosecution policies for domestic violence: A critical literature review and the case for more research to test victim empowerment approaches. Criminal Justice and Behavior, 25(3), 306–319.

National Institute of Justice (1997). Solicitation for the evaluation of Arrest Policies Program under the Violence Against Women Act. Retrieved November 15, 1998 from the World Wide Web: http://www.ncjrs.org/arrest3.txt.

Neighbors, H., & Jackson, J. (1984). The use of informal and formal help: Four patterns of illness behavior in the black community. American Journal of Community Psychology, 12(6), 629–644.

Poelzer, D., & Poelzer, I. (1986). In our own words: Northern Saskatchewan Metis women speak out. Saskatoon, Saskatchewan, Canada: Lindenblatt & Hamonic.

Rand, M. (1997). Violence-related injuries treated in hospital emer- gency departments. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

Reidy, R., & Von Korff, M. (1991). Is battered women’s help seek- ing connected to the level of their abuse? Public Health Reports, 106(4), 360–365.

Schmidt, J., & Sherman, L. (1993). Does arrest deter domestic violence? American Behavioral Scientist, 36(5), 601–610.

Simon, L. (1995). A therapeutic jurisprudence approach to the legal processing of domestic violence cases. Psychology, Public Policy, and Law, 1(1), 43–79.

Sullivan, C., & Bybee, D. (1999). Reducing violence using community- based advocacy for women with abusive partners. Journal of Con- sulting and Clinical Psychology, 67(1), 43–53.

Sussman, L., Robins, L., & Earls, F. (1987). Treatment-seeking for depression by black and white Americans. . .. Social Science Medicine, 24(3), 187–196.

The Urban Institute (1995–1996, Winter). New efforts to combat do- mestic violence. Retrieved March 7, 1997 from the World Wide Web: http://www.urban.org/periodcl/prr25 3a.html.

Tutty, L., Bidgood, B., & Rothery, M. (1993). Support groups for bat- tered women: Research on their efficacy. Journal of Family Vio- lence, 8(4), 325–343.

Violence Policy Center (1999). When men murder women: An analysis of 1997 homicide data. Washington, DC: Author. Retrieved October 7, 1999 from the World Wide Web: http://www.vpc.org/studies/dv211a.htm.

Walker, L. (1999). Psychology and domestic violence around the world. American Psychologist, 54(1), 21–29.

Wattendorf, G. (1996). Prosecuting cases without victim cooperation. FBI Law Enforcement Bulletin, 65(4), 18–21.

Springer