Does Type of Child Risk Affect Whether Mothers Seek Assistance for Intimate Partner Violence From Civil or Criminal Court?

Abstract

We examined whether risks to children of intimate partner violence survivors affected the type of legal assistance accessed. We hypothesized that the level and type of perceived child risk would be associated with whether women sought a protection order in civil court or filed charges against a current or former intimate partner in criminal court. Using data from a sample of predominantly African American women (N=293), we found that some forms of child risk were positively associated with seeking a civil order of protection but negatively associated with pressing criminal charges. Implications for practice, policy, and future research are presented.

Keywords

domestic violence protective order help-seeking children criminal justice system women African American

Intimate partner violence (IPV) is a significant public health problem with devastating consequences for women (El-Mouelhy, 2004).¹ Data from the National Intimate Partner and Sexual Violence Survey indicated that, on average, 35.6% of women in the United States have experienced rape, physical violence, and/or stalking by an intimate partner at some point in their lifetime (Black et al., 2011). This study also found that in the United States, African American, Latina, Multiracial, and Native women experience higher than average rates of IPV (Black et al., 2011). IPV victimization can have significant, long-term negative effects on the mental and physical health of women and their children (Bonomi et al., 2009; Holt, Buckley, & Whelan, 2008; Jordan, Campbell, & Follingstad, 2010).

For victims of IPV, concern for the safety and well-being of children is often a primary motivation for seeking help (Burgess-Proctor, 2008; Kelly, 2009; Khng & Ow, 2009; Petersen, Moracco, Goldstein, & Clark, 2005; Ptacek, 1999; Taylor, 2002). Therefore, the role that children play in women’s service utilization and decision-making processes is an important area of empirical research (Khaw & Hardesty, 2007). Specifically, research on the role that risks to child well-being play in motivating mothers to seek or utilize formal assistance through the court system is limited. Thus, the purpose of our study was to investigate whether risks to children were associated with where women received assistance within court systems. We focused on a sample of primarily African American women² who experienced IPV and sought formal assistance from civil and/or criminal court systems in an urban setting.

The framework for this study is based on Gondolf’s (1988) survivor hypothesis, which asserts that women who experience IPV are “active survivors” who “seek assistance in proportion to the realization that they and their children are more and more in danger” (Gondolf, 1988, p. 18). The current study tests this hypothesis by examining the extent to which participants sought different types of assistance in proportion to the level of risk to their children (Gondolf, 1988).

Background Literature

IPV Victimization and Involvement in the Court System

The decision to seek help due to violence is not simple. The process of help-seeking involves a multi-layered risk-benefit analysis to determine whether to seek help, where to access help, to whom to disclose the abuse, and whether to leave an abusive partner (Kelly, 2009; Liang, Goodman, Tummala-Narra, & Weintraub, 2005). The risks associated with seeking help include the possibility that the family will not be able to stay intact and the existence of stigma and shame around disclosure of abuse, whereas the benefits include the possibility of increasing physical safety and receiving the needed support (Kelly, 2009). Women who place priority on their role as mother must resolve many decisions when considering outside help, such as whether to remain in or leave the relationship for the sake of their children, all while trying to protect their children from experiencing abuse or witnessing their own suffering (Kelly, 2009).

The U.S. legal system provides both civil and criminal options for women who experience IPV. For example, in the civil court system, a woman can apply for an order of protection³ that mandates that the abuser stay away from her and her children (Moe, 2007). Meanwhile, in the criminal court system, a woman can cooperate in the criminal prosecution of an abuser. If convicted, an abuser’s sentence could include mandated treatment, probation, or even jail time (Moe, 2007). Although these protections exist, women face many barriers when they attempt to access assistance from a court system, including that the courts offer little control to the victim once the process has been initiated (Jordan, 2004).

Women who experience IPV victimization typically want one thing: for the abuse to stop (Burgess-Proctor, 2008). Yet, women’s reasons for seeking assistance through the court systems are somewhat mixed. Patzel (2001) found that women’s actions to seek help or end the relationship were positively associated with an escalation of violence. Similarly, other researchers have found that women sought protective orders when the violence became more severe (Duterte et al., 2008; Goodman, Dutton, Weinfurt, & Cook, 2003; Jordan, 2004), more frequent (Fischer & Rose, 1995), or as a last resort (Zoellner et al., 2000). Longer durations of physical and sexual IPV have also been significantly associated with women’s use of civil protective orders (Duterte et al., 2008). For women who do leave abusive relationships, it often takes them multiple attempts before they are successful (Bell, Goodman, & Dutton, 2007; Khaw & Hardesty, 2007; Taylor, 2002), and ending a relationship does not necessarily stop the abuse (Bell et al., 2007).

Some researchers have examined the characteristics of women who seek civil orders of protection. For example, Linares et al. (1999) found that women seeking a civil protective order experienced more verbal aggression and physical violence and had poorer physical and mental health compared with women who experienced IPV but did not seek protective orders. Wolf, Holt, Kernic, and Rivara (2000) found that women who sought civil protection orders were more likely to be employed and married, and less likely to be involved with the abuser or have experienced physical injury due to the incident that brought them to court.

Although some women may feel more in control of their situation after prosecution in criminal court (Ford & Regoli, 1992), dealing with the criminal justice system can leave women feeling revictimized (Letourneau, Duffy, & Duffett-Leger, 2012). It is possible that some women prefer to bring civil action rather than criminal action because they feel they will have more control over the outcome in civil court (Buzawa, Hotaling, & Klein, 1998). Therefore, the decision to seek assistance from the legal system is complex and multi-faceted, and once a woman decides to seek help, her options of where to seek help are as equally complex.

Factors That Deter Women’s Formal Help-Seeking

Many women who experience IPV encounter barriers that prevent them from seeking help from formal systems, including low self-esteem (Petersen et al., 2005), shame or fear of being blamed (E. Park, 2009), lack of awareness of potential sources for help (Fugate, Landis, Riordan, Naureckas, & Engel, 2005; E. Park, 2009), an inability to admit the severity of the abuse (E. Park, 2009; Petersen et al., 2005), fear of reprisal from the abuser (Fugate et al., 2005; Jordan, 2004; Taylor, 2002), fear of losing their children (Kelly, 2009; E. Park, 2009), or a need to protect the abuser or family members (Fugate et al., 2005; E. Park, 2009; Petersen et al., 2005). Some women also may face familial or social pressures to stay with the abuser, especially if they are married (Khaw & Hardesty, 2007).

Additional barriers exist for African American women, including a lack of accessible or culturally sensitive services and the fear of discrimination and institutional racism within the legal system (Bent-Goodley, 2004; Crenshaw, 1991; Richie, 1996; Sokoloff & Dupont, 2005; West, 2004) combined with “a community code of silence and the façade of the strong Black woman who should be able to survive without assistance” (Taylor, 2002, p. 81). In a qualitative study of 21 African American women who experienced IPV, Taylor (2002) reported that one woman experienced open hostility and rejection from her family when she left an abusive relationship. In addition, several women in this study reported receiving “negative and unhelpful responses from relatives, police and clergy” (Taylor, 2002, p. 80).

Motherhood and Women’s Involvement in the Court System

For many victims who are mothers, help-seeking and service utilization result from concerns over their children’s safety and well-being (Davis, 2002; Kelly, 2009; Moe, 2007; Patzel, 2001; Taylor, 2002). However, the presence of children can both facilitate and hinder women’s desire and ability to leave an abusive relationship (Kelly, 2009; Rhodes, Cerulli, Dichter, Kothari, & Barg, 2010). Women who experienced IPV were more likely to seek help from police when children were in the home (Bonomi, Holt, Martin, & Thompson, 2006), and this was especially true when their children witnessed the violence (Akers & Kaukinen, 2009; Meyer, 2010). Rhodes et al. (2010) conducted focus groups with 39 women and found that children motivated some of the women to call the police or assist the prosecutor to convict the abuser, but that children were also a factor in women’s decisions to remain in abusive relationships in an effort to keep the family together. Some women have reported waiting to leave their abusive partner until their children were grown in an effort to maintain the family unit, which was a decision they later regretted (Kelly, 2009).

Many mothers are active help-seekers when it comes to protecting their children, but they are well aware that attempting to leave or disclosing the abuse to formal systems often places women and children at risk for an escalation of violence (Bell, Goodman, & Dutton, 2009; Jaffe, Crooks, & Bala, 2009; Kelly, 2009; Moe, 2007; Taylor, 2002). Therefore, although it seems that women with children would be more likely to utilize the court system as a means of attempting to stop the abuse, the literature is mixed. Compared with women without children, women with children are more likely to participate in both criminal and civil interventions, and women who have children in common with the abuser are slightly more likely to seek a civil order of protection than women without children in common (Rhodes, Dichter, Kothari, Marcus, & Cerulli, 2011). Zoellner et al. (2000) found that threats by a woman’s partner to kill her were associated with her obtaining a restraining order, yet threats to her children were a deterrent.

Women who have children in common with the abuser have a greater risk of revictimization (Carlson, Harris, & Holden, 1999) and face legal and social challenges due to familial or social pressure (Burgess-Proctor, 2008; Kelly, 2009; Khaw & Hardesty, 2007; Taylor, 2002). Furthermore, if the abuser is legally the father of their children, he has the same legal right to their children as she does. Women who share children with an abuser can never completely separate from him if he retains and exercises custody and/or visitation rights (Taylor, 2002). Despite legal evidence of past abusive behavior by the father, some women lose custody of their children to abusive former partners (Saunders, 2007), and some men who perpetrate IPV utilize family court as an additional tactic of abuse and control over the victim (Jaffe et al., 2009). Some women may choose not to proceed with cooperating in the prosecution of their abusive partner out of concerns for their children’s well-being (Hare, 2010; Rhodes et al., 2010).

In general, however, women seek help to protect their children from the abuse or to prevent the abuser from taking the children from her (i.e., illegally through kidnapping or legally through gaining custody) (Bennett, Goodman, & Dutton, 1999; Goodman et al., 2003; E. Park, 2009; Ptacek, 1999). For some women, accessing formal help puts them in a position where they risk losing their children (Kelly, 2009; Moe, 2007; E. Park, 2009). In a study of women seeking help from Massachusetts courts, the majority of women reported that the abuser threatened to take their children, take her to court for custody, or contact Child Protective Services (CPS) and claim she was an unfit mother (Ptacek, 1999). Disclosing violence within a formal system can mean that CPS becomes involved, which could lead to removal of children from the home (Haight, Shim, Linn, & Swinford, 2007; Kelly, 2009). In Kelly’s (2009) qualitative study with 17 immigrant Latina women, she found that a fear of being separated from their children prevented the women from seeking help or leaving an abusive relationship.

Despite these challenges, the decision-making process for many women who experience IPV remains child-focused (Khaw & Hardesty, 2007). Along with other cognitive appraisals that go into the help-seeking process, mothers often make decisions “about their safety and that of their children in the context of fear, danger and inequity in and outside their homes” (Kelly, 2009, p. 288).

The Current Study

Few researchers have examined the role of child risk in which court system women who have experienced IPV access for help. Based on the empirical literature reviewed above, we formulated our hypothesis about whether child risk affects which court system a woman accesses for help. Our hypothesis was two-tailed due to the mixed evidence about the role children play in women’s help-seeking trajectories (Kelly, 2009; Rhodes et al., 2010). That is, we do not specify in our hypothesis whether risk of harm to children would be positively or negatively associated with seeking assistance from civil or criminal court. However, after controlling for demographic and situational variables, we expected women’s perceived risks of harm to their children would be associated with differential court involvement.

Method

Data Source

The data used in this secondary analysis were from a National Institute of Justice-funded longitudinal study of women who sought help for IPV (see Dutton, Goodman, Lennig, Murphy, & Kaltman, 2005). The purpose of the larger study was to examine the longitudinal course of the impact of IPV victimization among predominately African American women who sought help from one of three sources (shelter, civil protection order court, and criminal domestic violence court). Researchers from George Washington University, Georgetown University, University of Maryland, and Boston College recruited and either conducted in-person interviews or provided surveys to IPV survivors across 10 distinct points of contact between 1999 and 2004. Women were first recruited to participate at one of three sites in a major city on the eastern seaboard: District Court Civil Division (hereafter civil court), District Court Domestic Violence Criminal Docket (hereafter criminal court), and a shelter for abused women and their children. Eligibility criteria included experiencing abuse perpetrated by a male who was a current or former intimate partner, English speaking, sober and without significantly impaired mental status at the time of the initial interview. The initial baseline interview and/or questionnaire required approximately 45-60 min to complete and participants were paid US$20 for their time. Research assistants approached 522 women in total and 116 refused to participate (response rate = 73.4%). A number of reasons were given for refusing to participate, but most were related to time constraints. The baseline data were used for the current study.

Analytic Sample

The focal participants of the current study were women with children who were seeking legal assistance from civil or criminal court. The recruitment sites used in the larger study offered different types of legal assistance. For example, the women seeking help from civil court were filing for a temporary order of protection whereas the women seeking assistance from criminal court were recruited after a final disposition in a misdemeanor criminal case against an abusive partner. The current study focused on the first point of contact and only included women who had children and women for whom we had complete data. In the original data set, 368 of 406 participants reported having one or more children. Of the remaining 368 women, an additional 75 cases were excluded from this study’s sample due to missing data.⁴ Therefore, the analytic sample for the current study included 293 women who had at least one child.

Measures

Dependent variables

There were two dependent variables in this study informed by the following: (a) the site where the woman was recruited for the study (civil or criminal court) and (b) whether she indicated in the survey that she had “filed a petition for an ex-parte/protection order” in civil court or had “filed or tried to file criminal charges with the Court Commissioner against him.” Two dichotomous variables were created to represent whether a woman had sought an order of protection from civil court (1 = yes; 0 = no) and whether she had filed charges in criminal court (1 = yes; 0 = no).

Independent variables

To measure the independent variables, child risks, we relied on three survey questions where respondents were asked to predict how likely their abusive partner would engage in specific behaviors in the next 12 months. In the survey, women were asked, “In the next year, how likely do you think it is that he will” (a) “try to take away or kidnap your child or children?” (risk of kidnapping); (b) “try to get custody of your child or children?” (risk of custody loss); or (c) “cause emotional harm to your child or children?” (risk of emotional harm). Response options for the 3 questions were on a 5-point Likert-type scale (1 = low risk, 2 = medium low risk, 3 = medium risk, 4 = medium-high risk, and 5 = high risk). The alpha coefficient for these items was .81 in the current study sample. For analytic purposes, responses to the three items were recoded into three dichotomous variables. Children were considered “at-risk” if the mother rated the level of risk as four or higher on the item. In the analyses, we included each of the three types of risk separately. We also combined these three items into a fourth overall child risk variable that indicated whether the respondent rated any child risk item as four or higher.

Control variables

Based on the prior empirical research presented above, we included several control variables in the study. To measure the presence of children in the home, we used the response provided to the question, “How many of your children live with you?” This variable was transformed into a dichotomous variable that indicated whether the number of children currently living in the home was above or below the mean (1 = two or more children; 0 = fewer than two children). To measure whether each woman had children in common with the abuser, we used women’s responses to the question “How many children do you have with the man you are here about today?” This variable was also transformed into a dichotomous variable that indicated whether she had children in common with the abuser (1 = yes; 0 = no).

To measure the severity of the risk of future IPV as perceived by women, a modified version of the Danger Assessment Scale (Campbell, 1995) was used. This scale measured the extent to which each woman was at risk of a domestic violence-related homicide or serious injury (Dutton et al., 2005). The scale included 12 questions, adapted from Campbell’s (1995) 15-item Danger Assessment Scale, each with response options of “yes” or “no.” Questions included, but were not limited to “Has the violence increased in severity during the past year?” “Has he ever threatened to kill you?” and “As far as you know, does he have access to a gun?” The scale score was calculated by summing all 12 responses and dividing by the total number of scale items. The alpha coefficient for this scale was .69 in our study, which is typical for this scale when studies have small samples (Campbell, 2007).

We also included several demographic characteristics as control variables. These included employment status, the length of the relationship with the abusive partner and dummy variables to indicate relationship status at the time of the incident. In the survey, women were asked, “What is your job situation?” and were able to select options from various categories that applied to them. For our study, responses were transformed to a dichotomous variable that represented whether they had paid employment (1 = working full- or part-time; 0 = not employed or not working for pay). Women were also asked, “For how long were you, or have you been, involved with this man?” This was included in the analyses as the number of years study participants were in a relationship with the abusive partner. To measure the status of the relationship, participants were asked, “At the time of the events that brought you here, what was your relationship with the man you are here about?” Women in the study selected from a list of seven relationship categories. In this study, we included dummy variables to indicate three types of relationships (married: 1 = yes, 0 = no; dating: 1 = yes, 0 = no; or separated/divorced: 1 = yes, 0 = no). In the analyses, being married was the reference category.

Analysis Strategies

To examine the association between our dependent variables and the categorical control variables (i.e., race/ethnicity, education level, employment status, federal or state assistance, relationship at the time of the incident, and living arrangement), we used chi-square tests. To examine the association between our dependent variables and the continuous variables (i.e., danger assessment score, age, number of children in the home, and number of shared children), we used bivariate regression. Next, to test our hypothesis and examine the relationship between risk to child(ren) and civil or criminal court system involvement while controlling for other relevant variables, we conducted two types of multivariate analyses: logistic regression and seemingly unrelated bivariate probit. Because we hypothesized that each child risk variable may differentially affect women’s involvement in civil and criminal courts, several models were run. First, we conducted eight separate logistic regression analyses to examine the relationship between each child risk variable and each court option. Second, because involvement with criminal or civil court may influence involvement with the other, we conducted four separate analyses using seemingly unrelated bivariate probit (Long, 1997). This method allows for the estimation of two equations simultaneously (H. M. Park, 2009), which improves the efficiency of the estimators (Cameron & Trivedi, 2010).

The first two logistic regression models and the first bivariate probit included the overall child risk variable. The third and fourth logistic regressions and the second bivariate probit model focused on risk of kidnapping. The fifth and sixth logistic regressions and the third bivariate probit model included risk of custody loss. Finally, the seventh and eight logistic regressions and the fourth bivariate probit model included the variable for risk of emotional harm. All analyses were performed using Stata 11.2 (StataCorp, 2009).⁵

Results

Descriptive Statistics

Due to the recruitment strategy of the original study, all women in the sample were seeking assistance from a formal source (i.e., civil court, criminal court, domestic violence shelter) or a combination of these sources. The majority (67.9%) were seeking or already had a civil order of protection at the time of their recruitment in the study and half (49.8%) had been involved in the criminal prosecution of the abuser. These two categories are not mutually exclusive as 27% of the women were seeking help from both civil and criminal court.

As displayed in Table 1, the majority of the sample (80.2%) was African American. The study participants were, on average, 31.6 years old (range = 18-57). They had, on average, two children living with them (range = 0-6) and, on average, shared one child with the abuser. More than half (55.6%) had a high school diploma or less, 38.6% had attended some college or received an associate’s degree, and 5.8% had a bachelor’s degree or higher. The majority of women in the sample were employed either full-time (46.5%) or part-time (12.5%). Less than half of the sample (43.3%) was receiving federal or state assistance at the time of the survey. The women in the study had various relationship and living situations. The following categories of relationship existed at the time of the domestic violence incident. Some were in a current relationship (i.e., the man was a boyfriend, 39.5%; or husband, 27%), while others were separated (i.e., ex-boyfriend, 21.2%; separated, 9.9%; or divorced, 2.4%). More than half were living with him at the time of the incident (58.7%), while 30% were not living together and 11.3% reported that they were living with one another “on and off.”

Table 1.

Demographic Characteristics of Sample (N = 293).

Variable	%	M (SD)	Range
Race/ethnicity
African American	80.2
White	14.6
Other	5.2
Age		31.6 (7.7)	18-57
Number of children living in home		1.8 (1.1)	0-6
Number of shared children		0.9 (1.0)	0-6
Education level
High school diploma or less	55.6
Trade school, some college, associate’s degree	38.6
Bachelor’s degree or grad school	5.8
Employment status
Unemployed	29.6
Employed full-time	46.5
Employed part-time	12.5
Student	9.4
Military	1.0
Retired	1.0
Federal or state assistance	43.3
Relationship with man at the time of the incident
Married	27.0
Separated	9.9
Divorced	2.4
Boyfriend	39.5
Ex-boyfriend	21.2
Living arrangement
Living together	58.7
Living together on and off	11.3
Not living together	30.0

As for the level of child risk, approximately one fifth to one fourth of women in the sample rated the risk of kidnapping (22.5%), custody loss (24.9%), or emotional harm (25.6%) in the following year to be medium-high to high risk and 38.6% of respondents rated at least one child risk item as medium-high or high risk. When asked about the extent of IPV, a majority of the women reported experiencing injury (78.7%). A majority of the women also reported they had experienced stalking (85.9%) and almost half (47.7%) reported experiencing sexual violence by a current or former intimate partner. To put this into context, the five most endorsed items on the Danger Assessment scale indicate the high levels of risk and violence the women in the study were facing. They were as follows: currently separated or discussing separation (84.1%), she thinks he is capable of killing her (71.2%), he has threatened to kill her (67%), he is violently and constantly jealous of her (61.6%), and violence or abuse has increased in frequency in the past year (58.6%).

Bivariate Analyses

As a preliminary step, we examined the bivariate relationship between the 12 variables (including the 4 child risk variables and 8 control variables) and each dependent variable (civil and criminal court). The results (not shown) revealed that each independent and control variable (except having children in common with the abuser) was significantly associated with at least one of the dependent variables (p < .05). The overall child risk, risk of emotional harm, number of years involved with the abuser, and the relationship status of dating at the time of the incident were associated with seeking assistance from civil and criminal court. Additional variables that were associated with seeking help at civil but not criminal court were presence of children in the home, overall Danger Assessment score, being employed, and being separated or divorced from the abuser at the time of the incident. The variables that were associated with seeking help at criminal but not civil court were risk of kidnapping, risk of custody loss, and being married to the abuser at the time of the incident. We retained all the independent and control variables in the multivariate analyses.

Logistic Regression Analyses

As displayed in Table 2, the overall results of the multivariate logistic regression models indicated that for overall child risk (i.e., women who indicated that any child risk item was at medium-high or high risk), the odds of seeking help from civil court increased (odds ratio [OR] = 1.98; p = .038), whereas the odds of seeking help from criminal court decreased (OR = 0.53; p = .026; see Models 1 and 2 in Table 2).

Table 2.

Logistic Regression Results (N = 293).

Variable	Overall child risk (Models 1 and 2)				Risk of kidnapping (Models 3 and 4)				Risk of custody loss (Models 5 and 6)				Risk of emotional harm (Models 7 and 8)
	Civil court		Criminal court		Civil court		Criminal court		Civil court		Criminal court		Civil court		Criminal court
	Odds ratio	b (SE)	Odds ratio	b (SE)	Odds ratio	b (SE)	Odds ratio	b (SE)	Odds ratio	b (SE)	Odds ratio	b (SE)	Odds ratio	b (SE)	Odds ratio	b (SE)
Overall child risk	1.98*	0.68 (0.65)	0.53*	−0.63 (0.15)	–	–	–	–	–	–	–	–	–	–	–	–
Risk of kidnapping	–	–	–	–	1.43	0.36 (0.54)	0.62	−0.48 (0.21)	–	–	–	–	–	–	–	–
Risk of custody loss	–	–	–	–	–	–	–	–	1.49	0.40 (0.53)	0.68	−0.39 (0.21)	–	–	–	–
Risk of emotional harm	–	–	–	–	–	–	–	–	–	–	–	–	2.36*	0.86 (0.87)	0.69	−0.38 (0.20)
Lives with two or more children	2.03*	0.71 (0.55)	1.22	0.20 (0.30)	2.09**	0.73 (0.57)	1.14	0.13 (0.28)	2.07**	0.73 (0.56)	1.17	0.16 (0.29)	2.04*	0.71 (0.56)	1.20	0.18 (0.30)
Shares children with the abuser	0.47*	−0.75 (0.15)	1.06	0.06 (0.30)	0.53	−0.64 (0.17)	1.03	0.03 (0.30)	0.51*	−0.66 (0.17)	1.01	0.01 (0.30)	0.52*	−0.66 (0.16)	0.91	−0.09 (0.25)
Danger assessment score	3.27	1.18 (2.23)	0.75	−0.28 (0.47)	4.24*	1.45 (2.87)	0.64	−0.45 (0.39)	4.46*	1.50 (2.95)	0.57	−0.56 (0.34)	3.38	1.22 (2.26)	0.61	−0.49 (0.37)
Employed	2.56**	0.94 (0.71)	0.97	−0.03 (0.24)	2.49**	0.91 (0.68)	0.98	−0.02 (0.24)	2.49**	0.91 (0.69)	0.98	−0.02 (0.24)	2.48**	0.91 (0.69)	0.99	−0.01 (0.25)
Number of years with the abuser	1.05	0.05 (0.30)	0.95*	−0.06 (0.02)	1.05	0.05 (0.03)	0.95*	−0.06 (0.02)	1.05	0.05 (0.03)	0.95*	−0.05 (0.02)	1.05	0.05 (0.03)	0.95*	−0.05 (0.02)
Relationship status^a
Dating	0.60	−0.52 (0.21)	1.90*	0.64 (0.61)	0.58	−0.55 (0.20)	1.97*	0.68 (0.62)	0.58	−0.54 (0.20)	1.98*	0.68 (0.62)	0.61	−0.49 (0.22)	1.93*	0.66 (0.61)
Separated or divorced	–	0.24 (0.47)	1.38	0.33 (0.44)	1.26	0.23 (0.46)	1.41	0.34 (0.45)	1.29	0.26 (0.48)	1.37	0.32 (0.43)	1.27	0.24 (0.47)	1.39	0.33 (0.44)
Hosmer–Lemeshow goodness–of–fit test (p value)	.33		.48		.09		.44		.14		.10		.38		.08

Note. Odds ratios are presented with the coefficient (b) and standard errors (SE). Dash in a cell indicates that the variable was not included in the model (e.g., Models 1 and 2 did not include risk of kidnapping, risk of custody loss or risk of emotional harm variables).

Reference category: Married.

p < .05. **p < .01. ***p < .001.

Of the control variables included in Models 1 and 2, the presence of two or more children in the home was positively associated with seeking assistance at civil court (OR = 2.03; p = .010), whereas having a child(ren) in common with the abuser was negatively associated with seeking help through civil court (OR = 0.47; p = .022). Also, for women who were employed, relative to women who were unemployed or not working for pay, the odds of seeking assistance from civil court increased (OR = 2.56; p = .001). These variables were not associated with seeking assistance from criminal court. However, the number of years involved with the abuser was negatively associated with seeking assistance from criminal court (OR = 0.95; p = .026). Also, compared with women who were married at the time of the incident, the odds of seeking assistance from criminal court increased for women who were dating the abuser (OR = 1.90; p = .043).

The risk of kidnapping (Models 3 and 4 in Table 2) was not associated with seeking assistance from either court nor was the risk of custody loss (see Models 5 and 6 in Table 2). However, for both the risk of kidnapping and the risk of custody loss models, the presence of children in the home (OR = 2.09, p = .007; OR = 2.07, p = .008, respectively), the level of danger (OR = 4.24, p = .033; OR = 4.46; p = .024, respectively) and being employed (OR = 2.49, p = .001; OR = 2.49, p = .001, respectively) were positively associated with seeking assistance from civil court. For women who shared children with the abuser, there was a decrease in the odds of seeking help from civil court when custody loss was a concern (OR = 0.51; p = .044; see Model 5 in Table 2). Meanwhile, the number of years involved with the abusive partner was negatively associated with criminal court (OR = 0.95; p = .028), and dating the abuser (compared with being married to the abuser) was positively associated with seeking assistance from criminal court (OR = 1.97; p = .032).

The results of the risk of emotional harm analysis (see Models 7 and 8 in Table 2) were similar to the overall child risk results (presented above). For women who indicated there was a medium-high or high risk of emotional harm to the child, the odds of seeking assistance from civil court increased (OR = 2.36; p = .020; see Model 7 in Table 2). Having two or more children in the home was positively associated with seeking assistance from civil court (OR = 2.04; p = .010), whereas sharing a child(ren) with the abuser was negatively associated with seeking this type of assistance (OR = 0.52; p = .035). Also, employed women were more likely to be seeking assistance from civil court (OR = 2.48; p = .001).

The Hosmer and Lemeshow’s goodness-of-fit tests resulted in non-significant p values, which indicated good overall fit for each model (see Table 2, bottom row).

Bivariate Probit Analyses

Not surprisingly, accessing help from one type of court was found to be associated with seeking assistance from the other (χ² = 23.00; p < .001). Therefore, following the logistic regression analyses, we ran four models using seemingly unrelated bivariate probit (results not shown). The findings mirrored those of the logistic regression analyses. Overall child risk and risk of emotional harm were positively associated with seeking assistance from civil court, whereas overall child risk was negatively associated with seeking assistance from criminal court. Similar to the logistic regression results, there was no association between risk of kidnapping or risk of custody loss and accessing civil or criminal court in these analyses. The findings for the control variables also reflected those of the logistic regression analyses.

Predicted Probabilities

Next, we calculated the predicted probabilities of mothers choosing civil versus criminal based on the reported level of overall child risk and risk of emotional harm. As shown in Figure 1, the probability of mothers seeking assistance from civil court increased from .35-.52 as overall child risk increased. Alternatively, the probability of mothers seeking assistance from criminal court decreased from .26-.14, when overall child risk increased from low or medium risk to medium-high or high risk. There was a small change in probability of seeking neither civil nor criminal court (−.02) and both civil and criminal courts (.03).

Figure 1.

Predicted probabilities of court involvement by overall child risk.

Figure 2 presents the predicted probability of choosing civil versus criminal based on risk of emotional harm. The probability of mothers seeking assistance from civil court increased from .38-.52 as risk of emotional harm increased. Alternatively, the probability of mothers seeking assistance from criminal court decreased from .25-.13 as risk of emotional harm increased. Similar to the results for overall child risk, the change in probability of mothers seeking neither civil nor criminal court (−.04) and both civil and criminal court (.03) was small.

Figure 2.

Predicted probabilities of court involvement by risk of emotional harm.

Discussion

There are two factors that promote help-seeking: (a) recognizing a problem exists and (b) recognizing the problem will not subside without help from others (Cauce et al., 2002; Liang et al., 2005). In addition, particularly within marginalized communities (such as women of color who experience IPV), help-seeking does not typically occur until the problem is at a critical level, as defined by the survivor (Liang et al., 2005). The primary purpose of our study was to test the survivor hypothesis by examining whether child risk was associated with seeking help through the court system among mothers who experienced IPV. Relying on a sample of 293 mothers who predominantly identified as African American (80%), we explored whether risk of harm to children was positively or negatively associated with seeking assistance from civil or criminal court.

Our study hypothesis was partially supported. We found that women sought assistance from civil court over criminal court when they perceived their children to be in danger overall or at risk of emotional harm. We did not find an association between risk of custody loss or risk of kidnapping and involvement in either court system. This lack of association could be influenced by a number of factors, including the relative low base rate for endorsing high risk of kidnapping and custody loss. Furthermore, because most women were not married, nor in some cases living with the father of their child, the risk of custody loss may have played a less significant role.

Beyond our hypothesis, we found several control variables were associated with women’s involvement in civil and/or criminal court and these findings support the results of prior studies. Similar to findings by Wolf et al. (2000), we found that being employed was significantly associated with seeking help from civil court but not criminal court. In our study, the number of children in the home was positively associated with women’s help-seeking through civil court. This finding was similar to results found by Rhodes et al. (2011). However, contrary to other results presented by Rhodes et al., we did not find an association between the number of children in the home and women’s help-seeking through criminal court. We also found that the length of time a woman had been in a relationship with the abuser was negatively associated with seeking assistance from criminal court and that being in a dating relationship (compared with married) was positively associated with seeking assistance from criminal court. This latter finding is similar to the findings of a study of women of African descent in Baltimore and the U.S. Virgin Islands by Lucea et al. (2013).

Our analyses indicated that the severity of IPV was positively associated with women’s involvement in civil court and this finding was consistent with work by other researchers (Durfee & Messing, 2012; Duterte et al., 2008; Wolf et al., 2000). Yet, contrary to the findings of Dutton, Goodman, and Bennett (1999), we did not find a positive association between the severity of violence and women’s involvement in the prosecution of the abuser. It may be that seeking a civil order of protection is perceived to be a “safer” option than criminal court for some mothers. That is, a civil protective order can be a way the mother can protect herself and her children from the abuser (by requesting protection order remedies related to custody and visitation) without engaging in a process that involves criminal charges.

For some women who experience IPV, the criminal legal system is perceived as revictimizing, ineffective at protecting them, and involving unintended consequences (Barata, 2007; Epstein, Bell, & Goodman, 2003; Fugate et al., 2005). Women may be less likely to become involved with criminal court proceedings if they believe the prosecution of the abuser could lead to retaliation and even more severe violence. Feelings of fear over what the abuser might do, feeling frustration that the system moves too slowly and cannot adequately protect women, and feelings of guilt over possible incarceration of the abuser can all contribute to a woman not wanting to assist in the prosecution process (Bennett et al., 1999). These concerns could be viewed as “noncooperation” with the criminal justice system but women’s hesitancy to utilize criminal court could instead be seen as conscientious decisions to help maintain safety and protect themselves and their children (Fleury-Steiner, Bybee, Sullivan, Belknap, & Melton, 2006).

Based on a sample of predominantly African American women who experienced IPV, Fugate et al. (2005) found that among the reasons why they did not call the police were concerns about CPS involvement or not wanting to get the abuser arrested. African American women, in particular, may be reluctant to press criminal charges due to institutional racism toward them and the men in their lives (Burgess-Proctor, 2008; Taylor, 2002). It may also be that women are choosing to seek assistance from civil court, but not necessarily choosing to be involved with criminal court proceedings. That is, criminal court involvement could be the result of police involvement during an incident, whereas women who are seeking assistance from civil court may not have had a police-involved incident. This was supported by a previous finding that women preferred civil court over criminal court because they felt they had more control over the outcome (Buzawa et al., 1998).

Some women are reluctant to be involved with criminally prosecuting their abuser over concerns of incarceration and how this could limit their receipt of child support and children’s access to their father (Bennett et al., 1999). In our study, we did not find any association between sharing children with the abuser and seeking help through criminal court. We did, however, find a negative association between sharing children with the abuser and involvement in civil court. When women who share children with the abuser seek an order of protection, the court order may involve a visitation plan for the father to see the children (Sheeran & Hampton, 1999). If women know this through past experiences or word-of-mouth, it may be that their informal visitation arrangement (or lack of an arrangement) is preferred to one that would be mandated through the civil court.

Study Limitations

As with all research, our results must be considered along with the limitations of our study. One limitation was that all the participants were already seeking assistance from a formal court-based setting or a shelter when they were recruited for the study. We were not able to capture information on women’s decision-making process and explore why (or if) they chose to seek assistance through formal resources. In addition, the first point of contact measured in this study was not necessarily the first time women sought assistance through the courts. Yet, many low-income, African American women do not seek help due to a lack of available culturally competent services, a distrust of the criminal justice system or gender entrapment (Bent-Goodley, 2004; Richie, 1996). In this way, our findings shed some light on this under-recognized group of women. However, the generalizability of our findings is limited due to our sample composition of predominantly African American women seeking help in an urban environment.

A second limitation of this study was that we excluded 75 women from the original sample due to missing data. Most of these women (n = 55) were excluded because they did not answer the child risk questions. Due to our concern that the mothers who did not respond to these questions may differ from those who did respond to the child risk questions in the survey, we conducted independent group t tests with each of the independent variables. We found that mothers who did not answer the risk questions were less likely to have more than two children in the home and less likely to share children with the abuser. These differences were statistically significant (p < .001). Of the 55 mothers who did not respond to the child risk questions, 22 had no children living in the home and 20 had one child living in the home. Only 16% of the non-responders shared children with the abuser. There were no statistically significant differences among the two groups of women on any of the other independent variables. Therefore, it is possible that the majority of the non-responders did not provide information on our key predictor variables primarily due to inapplicability.

Because all women involved in this study were seeking help from a formal source, we were not able to examine how factors of child risk or other study variables might differ between women who are already seeking assistance through a formal resource (e.g., court, domestic violence shelter) and women who are seeking assistance through informal resources (e.g., family or friends). We were also unable to examine whether there were differences in the study variables depending on which other formal resources women used for assistance. Our study focused on women’s involvement with the court systems and we were not able to examine women’s use of mental health services or other formal resources. Researchers may wish to concentrate on these potential differences in future studies.

Conclusion

Despite the study limitations, our finding that women’s perceptions of child risk were positively associated with their involvement in civil court and negatively associated with criminal court has important implications. Although there has been some progress in this area in the past decade, civil and criminal court systems that serve victims of IPV must consider and screen for the level of risk their children are facing. Civil courts in particular are in a good position to develop remedies, such as court-mandated supervised visitation (where child transfer between separated parents is supervised), that address child risk directly. Researchers should continue to explore the extent that professionals within legal systems are screening for IPV and including remedies that protect children. Researchers should also investigate what information court personnel are using to inform visitation and child custody determinations. There has been some recent exploration of issues related to child custody and domestic violence (see M. S. Davis, O’Sullivan, Susser, & Fields, 2011; Saunders, Faller, & Tolman, 2011), but much more research in the area of child risk and help-seeking for IPV is needed. Researchers should consider including qualitative components to learn more about why women seeking help chose civil over criminal court (or vice versa).

As noted, some women prefer to bring civil action rather than criminal action (Buzawa et al., 1998) and we found that women’s perceptions of child risk increases the likelihood that they will become involved in the civil court system. Unfortunately, the civil court system also has its share of challenges. On average, civil orders of protection are violated 40% of the time (Spitzberg, 2002), and women with children are more likely than women without children to have their order of protection violated (Harrell & Smith, 1996; Logan, Shannon, Walker, & Faragher, 2006). It is our hope that the findings from this study can be utilized to better understand how children influence women’s decisions to seek assistance through the court systems and that research in this area continues to be conducted. Researchers should concentrate future efforts on learning more about the extent to which women are active survivors, by specifically focusing on how women experience the court systems and whether there can be procedural changes to better meet the needs of women who experience IPV and who are reaching out to the courts to improve the safety for their children and themselves.

Footnotes

Acknowledgements

The authors thank the study participants for their involvement in this research. They also thank the anonymous peer reviewers who provided thoughtful and helpful suggestions.

Authors’ Note

An earlier version of this article by the first author was awarded the Gerry T. Hotaling Memorial Student Research Award at the 2012 International Family Violence and Child Victimization Research Conference.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The third and fourth authors were the co-principal investigator and principal investigator, respectively, of the original study, which was supported by National Institute of Justice Grant 2001-WT-BX-0001.

Notes

Author Biographies

Jane E. Palmer, PhD, MSW, is a professorial lecturer in the Department of Public Administration and Policy in the School of Public Affairs at American University and the director of the Community-Based Research Scholars program. Previously, she was a post-doctoral associate at Rutgers University School of Social Work’s Center on Violence Against Women & Children and a graduate research assistant at the National Institute of Justice. Her research interests include help-seeking by survivors of gender-based violence from marginalized communities, policy responses to gender-based violence, the prevention of gender-based violence and methodological, measurement, and cultural competency issues in violence against women research.

Lynette M. Renner, PhD, MSW is an associate professor in the School of Social Work at the University of Minnesota. Her scholarship areas include understanding the prevalence and co-occurrence of family violence, identifying risk factors for specific types of family violence victimization and perpetration, and exploring associations between child maltreatment and intimate partner violence, and behavioral, educational, and mental health outcomes. Her recent research projects have focused on examining women’s help-seeking strategies, how victimization influences women’s parenting, and women’s perceptions of procedural justice.

Lisa A. Goodman, PhD, is a professor in the Department of Counseling and Applied Developmental Psychology at Boston College. She has authored more than 95 publications on the interaction between intimate partner violence survivors and the practices, systems, and policies that aim to help them; mental health practices for low-income communities; and bringing social justice to psychology. She consults extensively with local and national domestic violence organizations, and is the co-founder of the Massachusetts Domestic Violence Program Evaluation and Research Collaborative. Her 2008 book, co-authored with Deborah Epstein, is titled Listening to Battered Women: A Survivor-Centered Approach to Advocacy, Mental Health, and Justice.

Mary Ann Dutton, PhD, is a professor in the Department of Psychiatry at Georgetown University Medical Center and the co-director of the Community Engagement and Research component of the Georgetown Howard Universities Center for Clinical and Translational Science. She is also the associate director of the Georgetown Center for Trauma and the Community. Her research focuses on trauma, health, and mental health among low-income minority women and veterans. She has received numerous federal grants, including for longitudinal studies, randomized clinical trials involving both mindfulness and telehealth interventions, and immigrant women’s use of civil protection orders.

References

Akers

Kaukinen

(2009). The police reporting behavior of intimate partner violence victims. Journal of Family Violence, 24, 159-171.

Barata

P. C.

(2007). Abused women’s perspectives on the criminal justice system’s response to domestic violence. Psychology of Women Quarterly, 31, 202-215.

Bell

Goodman

Dutton

M. A.

(2007). The dynamics of staying and leaving: Implications for battered women’s emotional well-being and experiences of violence at the end of a year. Journal of Family Violence, 22, 413-428.

Bell

Goodman

Dutton

M. A.

(2009). Variations in help-seeking, battered women’s relationship course, emotional well-being, and experiences of abuse over time. Psychology of Women Quarterly, 33, 149-162.

Bennett

Goodman

Dutton

M. A.

(1999). Systemic obstacles to the criminal prosecution of a battering partner: A victim perspective. Journal of Interpersonal Violence, 14, 761-772.

Bent-Goodley

T. B.

(2004). Perceptions of domestic violence: A dialogue with African American women. Health & Social Work, 29, 307-316.

Black

M. C.

Basile

K. C.

Breiding

M. J.

Smith

S. G.

Walters

M. L.

Merrick

M. T.

. (2011). The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 Summary Report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.

Bonomi

A. E.

Anderson

M. L.

Reid

R. J.

Rivara

F. P.

Carrell

Thompson

R. S.

(2009). Medical and psychosocial diagnoses in women with a history of intimate partner violence. Archive of Internal Medicine, 169, 1692-1697.

Bonomi

A. E.

Holt

V. L.

Martin

D. P.

Thompson

R. S.

(2006). Severity of intimate partner violence and occurrence and frequency of police calls. Journal of Interpersonal Violence, 21, 1354-1364.

10.

Burgess-Proctor

(2008). Understanding the help-seeking decisions of marginalized battered women (Unpublished doctoral dissertation). Michigan State University, East Lansing.

11.

Buzawa

Hotaling

Klein

(1998). The response to domestic violence in a model court: Some initial findings and implications. Behavioral Sciences & the Law, 16, 185-206.

12.

Cameron

A. C.

Trivedi

P. K.

(2010). Microeconometrics using Stata. College Station, TX: Stata Press.

13.

Campbell

J. C.

(1995). Assessing dangerousness: Violence by sexual offenders, batterers and child abusers. Newbury Park, CA: Sage.

14.

Campbell

J. C.

(2007). Assessing dangerousness: Violence by batterers and child abusers. New York: Springer.

15.

Carlson

M. J.

Harris

S. D.

Holden

G. W.

(1999). Protective orders and domestic violence: Risk factors for re-abuse. Journal of Family Violence, 14, 205-226.

16.

Cauce

Domenech-Rodríguez

Paradise

Cochran

Shea

J. M.

Srebnik

Baydar

(2002). Cultural and contextual influences in mental health help seeking: A focus on ethnic minority youth. Journal of Consulting and Clinical Psychology, 70, 44-55.

17.

Centers for Disease Control and Prevention. (2013). Intimate partner violence: Definitions. Retrieved from http://www.cdc.gov/violenceprevention/intimatepartnerviolence/definitions.html

18.

Crenshaw

(1991). Mapping the margins: Intersectionality, identity politics, and violence against women of color. Stanford Law Review, 43, 1241-1299.

19.

Davis

M. S.

O’Sullivan

C. S.

Susser

Fields

M. D.

(2011). Custody evaluations when there are allegations of domestic violence: Practices, beliefs and recommendations of professional evaluators. Washington, DC: United States Department of Justice, National Institute of Justice.

20.

Davis

R. E.

(2002). Leave-taking experiences in the lives of abused women. Clinical Nursing Research, 11, 285-305.

21.

Durfee

Messing

J. T.

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

22.

Duterte

E. E.

Bonomi

A. E.

Kernic

M. A.

Schiff

M. A.

Thompson

R. S.

Rivara

F. P.

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

23.

Dutton

M. A.

Goodman

Lennig

Murphy

Kaltman

(2005). Ecological model of battered women’s experience over time: Final report. Washington, DC: U.S. Department of Justice, National Institute of Justice.

24.

Dutton

M. A.

Goodman

L. A.

Bennett

(1999). Court-involved battered women’s responses to violence: The role of psychological, physical, and sexual abuse. Violence and Victims, 14, 89-104.

25.

El-Mouelhy

(2004). Violence against women: A public health problem. Journal of Primary Prevention, 25, 289-303.

26.

Epstein

Bell

Goodman

(2003). Transforming aggressive prosecution policies: Prioritizing victims’ long-term safety in the prosecution of domestic violence cases. Journal of Gender, Social Policy & the Law, 11, 465-498.

27.

Fischer

Rose

(1995). When “enough is enough”: Battered women’s decision making around court orders of protection. Crime & Delinquency, 41, 414-429.

28.

Fleury-Steiner

R. E.

Bybee

Sullivan

C. M.

Belknap

Melton

H. C.

(2006). Contextual factors impacting battered women’s intentions to reuse the criminal legal system. Journal of Community Psychology, 34, 327-342.

29.

Ford

D. A.

Regoli

(1992). The preventive impact of policies for prosecuting wife batterers. In Buzawa

Buzawa

(Eds.), Domestic violence: The changing criminal justice response (pp. 181-207). Westport, CT: Greenwood.

30.

Fugate

Landis

Riordan

Naureckas

Engel

(2005). Barriers to domestic violence help seeking: Implications for intervention. Violence Against Women, 11, 290-310.

31.

Gondolf

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

32.

Goodman

L. A.

Dutton

M. A.

Weinfurt

Cook

(2003). The Intimate Partner Violence Strategies Index: Development and application. Violence Against Women, 9, 163-186.

33.

Haight

W. L.

Shim

W. S.

Linn

L. M.

Swinford

(2007). Mothers’ strategies for protecting children from batterers: The perspectives of battered women involved in Child Protective Services. Child Welfare, 86(4), 41-62.

34.

Hare

S. C.

(2010). Intimate partner violence: Victims’ opinions about going to trial. Journal of Family Violence, 25, 765-776.

35.

Harrell

Smith

(1996). Effects of restraining orders on domestic violence victims. In Buzawa

Buzawa

(Eds.), Do arrests and restraining orders work? (pp. 214-242). Thousand Oaks, CA: Sage.

36.

Holt

Buckley

Whelan

(2008). The impact of exposure to domestic violence on children and young people: A review of the literature. Child Abuse & Neglect, 32, 797-810.

37.

Jaffe

Crooks

Bala

(2009). A framework for addressing allegations of domestic violence in child custody disputes. Journal of Child Custody, 6, 169-188.

38.

Jordan

C. E.

(2004). Intimate partner violence and the justice system: An examination of the interface. Journal of Interpersonal Violence, 19, 1412-1434.

39.

Jordan

C. E.

Campbell

Follingstad

(2010). Violence and women’s mental health: The impact of physical, sexual, and psychological aggression. Annual Review of Clinical Psychology, 6, 607-628.

40.

Kelly

(2009). “I’m a mother first”: The influence of mothering in the decision-making processes of battered immigrant Latino women. Research in Nursing & Health, 32, 286-297.

41.

Khaw

Hardesty

J. L.

(2007). Theorizing the process of leaving: Turning points and trajectories in the stages of change. Family Relations, 56, 413-425.

42.

Khng

(2009). Getting out: Factors that influenced abused wives’ decisions to leave their abusive spousal relationships in Singapore. Families in Society, 90, 153-161.

43.

Letourneau

Duffy

Duffett-Leger

(2012). Mothers affected by domestic violence: Intersections and opportunities with the justice system. Journal of Family Violence, 27, 585-596.

44.

Liang

Goodman

Tummala-Narra

Weintraub

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

45.

Linares

L. O.

Groves

B. M.

Greenberg

Bronfman

Augustyn

Zuckerman

(1999). Restraining orders: A frequent marker of adverse maternal health. Pediatrics, 104, 249-257.

46.

Logan

T. K.

Shannon

Walker

Faragher

T. M.

(2006). Protective orders: Questions and conundrums. Trauma, Violence & Abuse, 7, 175-205.

47.

Long

J. S.

(1997). Regression models for categorical and limited dependent variables. Thousand Oaks, CA: Sage.

48.

Lucea

Stockman

Mana-Ay

Bertrand

Callwood

Coverston

. (2013). Factors influencing resource use by African American and African Caribbean women disclosing intimate partner violence. Journal of Interpersonal Violence, 28, 1617-1641.

49.

Meyer

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

50.

Moe

A. M.

(2007). Silenced voices and structured survival: Battered women’s help seeking. Violence Against Women, 13, 676-699.

51.

Park

(2009). Battered women’s help-seeking: A turning point from victimization to readiness (Unpublished doctoral dissertation). University of Minnesota, Minneapolis.

52.

Park

H. M.

(2009). Regression models for binary dependent variables using Stata, SAS, R, LIMDEP and SPSS (Working paper). Bloomington: Indiana University.

53.

Patzel

(2001). Women’s use of resources in leaving abusive relationships: A naturalistic inquiry. Issues in Mental Health Nursing, 22, 729-747.

54.

Petersen

Moracco

Goldstein

Clark

K. A.

(2005). Moving beyond disclosure: Women’s perspectives on barriers and motivators to seeking assistance for intimate partner violence. Women & Health, 40(3), 63-76.

55.

Ptacek

(1999). Battered women in the courtroom: The power of judicial responses. Boston, MA: Northeastern University Press.

56.

Rhodes

K. V.

Cerulli

Dichter

M. E.

Kothari

C. L.

Barg

F. K.

(2010). “I didn’t want to put them through that”: The influence of children on victim decision-making in intimate partner violence cases. Journal of Family Violence, 25, 485-493.

57.

Rhodes

K. V.

Dichter

M. E.

Kothari

C. L.

Marcus

S. C.

Cerulli

(2011). The impact of children on legal actions taken by women victims of intimate partner violence. Journal of Family Violence, 26, 355-364.

58.

Richie

(1996). Compelled to crime: The gender entrapment of Black battered women. New York: Routledge.

59.

Saunders

D. G.

(2007). Child custody and visitation decisions in domestic violence cases: Legal trends, risk factors, and safety concerns. Harrisburg, PA: National Online Resource Center on Violence Against Women. Retrieved from http://www.vawnet.org/applied-research-papers/print-document.php?doc_id=371

60.

Saunders

D. G.

Faller

K. C.

Tolman

R. M.

(2011). Child custody evaluators’ beliefs about domestic abuse allegations: Their relationship to evaluator demographics, background, domestic violence knowledge and custody-visitation recommendations. Washington, DC: U.S. Department of Justice, National Institute of Justice.

61.

Sheeran

Hampton

(1999). Supervised visitation in cases of domestic violence. Juvenile & Family Court Journal, 50, 13-25.

62.

Sokoloff

N. J.

Dupont

(2005). Domestic violence at the intersections of race, class and gender: Challenges and contributions to understanding violence against marginalized women in diverse communities. Violence Against Women, 11, 38-64.

63.

Spitzberg

(2002). The tactical topography of stalking victimization and management. Trauma, Violence & Abuse, 3, 261-288.

64.

StataCorp. (2009). Stata Statistical Software: Release 11. College Station, TX: StataCorp LP.

65.

Taylor

(2002). “The straw that broke the camel’s back”: African American women’s strategies for disengaging from abusive relationships. Women & Therapy, 25(3/4), 79-94.

66.

West

(2004). Black women and intimate partner violence: New directions for research. Journal of Interpersonal Violence, 19, 1487-1493.

67.

Wolf

M. E.

Holt

V. L.

Kernic

M. A.

Rivara

F. P.

(2000). Who gets protection orders for intimate partner violence? American Journal of Preventive Medicine, 19, 286-291.

68.

Zoellner

L. A.

Feeny

N. C.

Alvarez

Watlington

O’Neill

M. L.

Zager

. (2000). Factors associated with completion of the restraining order process in female victims of partner violence. Journal of Interpersonal Violence, 15, 1081-1099.