The authors wish to thank Sally McNichol of the Urban Justice Center as well as Susan Urban, Shirley Whitney, and Elaine Wolf of the New York City Administration for Children's Services who were instrumental in the development and implementation of this project.

Correspondence concerning this paper should be addressed to Randy H. Magen, Department of Social Work, University of Alaska-Anchorage, 3211 Providence Drive, Anchorage, Alaska 99508. Electronic mail may be sent via Internet to afrhm1@uaa.alaska.edu

In child welfare the term, "wrap around services" (McGowan & Meezan, 1983) is used to describe the continuum of services from prevention to family preservation to foster care and reunification programs. The few programs across the United States which have focused on the nexus of child maltreatment and domestic violence have been directed at a narrow segments of the child welfare continuum. For example, the AWAKE program in Boston works with maltreated children and battered women who are identified in a hospital setting (Schechter & Edleson, 1995). Other efforts have focused on child protective service investigations (Beszterczey, 1994; Magen, Conroy, Hess, Pancieria, & Simon, 1995). This paper reports on an effort to identify and serve battered women with children "at risk" of maltreatment--children who have entered the child welfare system at a time of less imminent danger than either a hospital admission or child maltreatment investigation. This is the first report from a child welfare setting that has focused on the coexistence of domestic violence and potential child maltreatment in a population of children "at risk".

Programs designed to assist families where children have been identified as being at risk of maltreatment are known within the child welfare system as preventive services. These are typically voluntary not-for profit, long-term, community-based or specialized programs which work to ameliorate the conditions which might lead to child maltreatment. In New York City preventative services are referred to as Purchased Preventive Services (PPRS) because the agencies operate under contract with the New York City Administration for Children's Services (ACS), the state designated child protective service agency for New York City. Families can be referred to PPRS agencies by ACS, by other sources, or families can make self-referrals. Annually, over 25,000 families receive preventive services in New York City (Citizens Committee for Children of New York, 1997). In 1996 about 30% of the PPRS caseload was comprised of referrals from ACS (Protecting the Children of New York, 1996).

Barriers to Serving Battered Women

The separate historical development and delivery of services in child welfare and services to battered woman has been compounded by different philosophies, different professional terminologies, and even the value placed on different outcomes (Schechter & Edleson, 1995). These differences have lead to a failure to recognize that efforts were often focused on the same family. Cummings and Mooney (1988) point out that while child welfare workers and battered women's advocates both "...share an interest in stopping the violence, their perspectives and approaches are frequently in conflict" (p. 4). Battered women's advocates usually adopt a woman-centered approach with the goal of empowering women. Child welfare workers, on the other hand typically adopt a child-centered approach and follow the principle of working "in the best interests' of the child."

Often the principle of "best interests of children" is translated into an assessment of who can keep the child safe and promote his or her well-being. A greater responsibility for protecting the child has usually been imposed on mothers than on fathers (Davis, 1995). Compounding this difficulty is the fact that the mother's own safety has typically not been assessed. Thus, in families where there was domestic violence, identification of the abuse was dependent upon the attitude, skill, and behavior of the individual caseworker. If a battered woman was lucky, she worked with a caseworker who did not blame her for the abuse. However a battered woman was likely to work with a caseworker oblivious to domestic violence and who viewed the battered woman's behavior as resistance which, in the caseworker's view, created an increased risk to the child(ren). In this type of casework the battered woman became caught between the batterer and the child welfare worker; a situation which could lead to the battered woman being doubly victimized, once by the batterer and a second time by the child welfare worker (Davis, 1995).

Writers from the Battered Women's Movement have also articulated the problem of simultaneously addressing of the issues of battered women and child abuse. Schechter (1982) pointed out several of these problems, from the reluctance of Battered Women's Movement founders to work with children--who comprise two thirds of the average shelter population--to activists working to keep wife abuse and child abuse separate on the federal funding and policy agenda. Schechter herself speaking at the Fifth National Conference on Child Abuse and Neglect in 1981 spoke "Against Consolidation with Child Abuse Services" stating "Child abuse and violence against women are different phenomena with different historical roots. A major goal of child protection services is to keep the family united, hopefully free from violence. Our goal is to support a woman in creating a violence-free life for herself and her children in whatever way she chooses." (1982, 324). Other writers such as Straus (1983) have argued that the need for immediate emergency intervention justifies separating services to battered women from assistance to abused children.

Addressing woman abuse in the context of child welfare preventive services is an ideal setting to abandon this dichotomy. In preventive services children are the identified family member at-risk; women and children, if not the family as a whole, are often the focus of treatment and intervention. Unlike protective services, preventive services does not have the stigma nor the role of child maltreatment investigation. Nor does preventive services have the focus or image of services for battered women. The umbrella label of "at-risk" connected with the child may enhance the readiness of battered woman to benefit from interventions.

The project reported in this paper sought to systematically change the manner in which child welfare preventive service agencies identified and worked with battered women. This undertaking involved training, case consultation, service development, and a new intake questionnaire. This paper reports on the use of the intake questionnaire.

The Family Violence Prevention Project

The intake questionnaire was a key component of the Family Violence Prevention Project (FVPP), a comprehensive program designed to address the co-existence of child and woman abuse in families that become known to the child welfare system. This project was initiated in 1993 through a collaboration of battered women's advocates, child welfare professionals, and University based social work researchers. The project arose out of a need for battered women's advocates and child welfare professionals to address the abuse of women and the maltreatment of children in a less bifurcated manner.

One of the basic beliefs underlying the project was that child welfare staff could and should offer interventions to abused women with the purpose of assisting in their efforts to protect themselves and their children. While this paper primarily discusses the intake questionnaire, it is important to realize that the assessment of domestic violence must take place in a context which tries, above all else, to enhance the safety for abused women and children. When battered women admit to being victims they risk, on one hand, involvement in the child protective system which can remove their children and on the other hand, a partner who can escalate the abuse; these risks can only be mitigated when the paramount goal is to maximize the safety of abused women and children.

The project assumed that domestic violence is a "hidden" force in many families which come to the attention of the child welfare system. Therefore, a basic component of the FVPP was to screen for domestic violence at intake on all families, regardless of the presenting problem. Due to the level of risk in domestically violent families a high level of awareness on the part of workers to issues of safety was necessary. Developing an understanding of the risks and the project philosophy required extensive amounts of training and support. The training and support also helped to insure that the intake questionnaire was utilized in a consistent manner. Although these philosophical and training components of the FVPP are beyond the purview of this paper, it is important for the authors to emphasize that intensive training and constant oversight of this work were essential for good intentions not to turn to bad programming which further endangered those we sought to help.

The Intake Questionnaire

The new intake questionnaire was developed through a collaborative effort between PPRS caseworkers and supervisors, battered womens' advocates, ACS social workers, and researchers. An existing CPS questionnaire (see Magen et al., 1995), research questionnaires (Straus, Gelles, & Steinmetz, 1981) and questionnaires utilized in other settings were the foundation for the new intake questionnaire. The purpose of the questionnaire was twofold; first, to provide caseworkers with a tool to screen for the presence of domestic violence; and two, to collect useful information about domestic violence in PPRS settings.

The final intake questionnaire had five sections: Face Sheet, Interview, Extent of Domestic Violence, Caseworker's Assessment, and Applicant's Evaluation (see Appendix). The cover sheet to the questionnaire offered a definition of domestic violence and instructed the caseworker to ask the domestic violence questions, "in private, without partner or other family members present." Caseworkers were asked to complete the questionnaire within 30 days of intake. Forms were provided to agencies in both English and Spanish.

The Face Sheet collected basic demographic information about people in the household including age, gender, legal relationship, race, religion, sources of income, and whether the woman's partner resided in the home.

The Interview Section had 10 questions which helped the caseworker move from asking about "normal" conflicts and arguments to actual abusive behavior. One of the first questions in the interview section was an expanded version of Straus, Gelles, & Steinmetz's (1981) marital conflict index. The original Marital Conflict Index consisted of five areas of conflict scored on a five point scale (1=always, 2=almost always, 3=usually, 4=sometimes, and 5=never). Six additional areas of conflict were added for the PPRS intake questionnaire; the original five point scale was maintained. Additional questions in the Interview Section asked about arguments, physical fights, threats, and fear of partner. One question which asked specifically about domestic violence read:

Domestic violence is a problem that affects many families in NYC of all different races, religions and incomes. Many women are abused by their intimate partners in emotional, sexual, physical, and materials ways. For example, some partners hit, push, punch, or kick. Others threaten them with harm, withhold money from them, force them to do things they don't want to do, or constantly criticize them. Do things like this happen in your relationship with your partner?

The next section of the questionnaire, Extent of Domestic Violence, was completed only when the interview revealed the presence of domestic violence. This section was designed to collect information on the type and frequency of abuse as well as efforts the woman had made to seek help.

The fourth section of the questionnaire asked the casework to make a domestic violence assessment related to the current as well as past family history. The caseworker was also asked to assess the frequency of the abuse and to describe what actions he/she was taking to help the client deal with the abuse.

Finally, before 120 days into the case, or at termination if occurring earlier than 120 days, the caseworker was asked to give the client the Applicant's Evaluation section. This section contained 6 consumer satisfaction questions which inquired as to how helpful it was to have been interviewed about domestic violence.

Seven PPRS agencies from 3 of New York's five Boroughs (Brooklyn, Manhattan, and Staten Island) agreed to participate in the first year of the project. Agencies were solicited for participation by ACS staff, participation was voluntary. The seven agencies agreed to implement the new intake questionnaire, participate in training and case consultation for one year, from June 1995 through May of 1996. In the second year of the project, nine additional agencies joined the project which expanded the coverage to all five of New York City's Boroughs. In addition to the training and case consultation, quarterly meetings were held between ACS staff, battered women's advocates, research staff, and agency representatives. These quarterly meetings were for information sharing, support, consultation, and to work out difficulties in implementing the domestic violence project.

There were four questions involved in the evaluation of the questionnaire: First, were more battered women identified using the new questionnaire than utilizing previous operating procedures? Second, what were the client's reactions to being asked about domestic violence in a child welfare agency? Third, what were the characteristics of the battered women population in this child welfare setting? Fourth, what unintended or secondary effects developed as a result of implementing the new intake questionnaire?

The demographic characteristics of the 540 families are reported in Table 1 and Table 2. Families in this sample had on average 4.2 people in the household (SD = 1.5). In only 170 (31.5%) of these families was the female client's partner reported as living in the household.

Table 1
Demographic Characteristics of Families (N= 540)

Characteristic		Woman	Partner	Children	Others in Household
N		540	328	1273	186
	M	n/a	n/a	2.4	.4
	SD	n/a	n/a	1.3	.9
Age
	M	33.6	36.3	8.2	30.2
	SD	8.5	10.8	4.7	20.1
Gender (N)
	Male	0	320	674	74
	Female	540	8	599	112
Source of Income (N)
	Employment	129	172
	Income Support	223	24
	Social Security	17	3
	SSI	37	15
	Other	100	43

Table 2
Demographic Characteristics of Female Client (N= 540)

Characteristic			N
Race
	African-American		158
	Asian		7
	Caucasian		67
	Hispanic		175
	Native American		2
	Other		12
Religion
	Catholic		172
	Jewish		10
	Muslim		10
	Protestant		50
	Other		73

Conflict

Previous national surveys of family violence have reported on the nature of conflicts in families as well as revealed a relationship between conflict and violence in the family (Straus et. al., 1981). The 10 areas of conflict on the intake questionnaire were analyzed in a similar manner. The overall reliability of the scale, as measured by Chronbach's alpha, was .86 and item-to-item correlations ranged from .42 for the item on pregnancy to .70 for the question on sexual relations.

To identify what the most common areas of conflict were in these families and for conceptual clarity, the scoring system for the questions was reversed (e.g."always" was coded 5 instead of 1). Next a mean score was generated and the resulting score was used to rank the conflictual issues argued about from the most common to least common. The higher the mean score the more often that source of conflict was argued about.

The five most common areas of conflict in these 540 families were: managing money (M = 2.55), things about children (M = 2.54), discipline of children (M = 2.48), extended family/relatives (M = 2.29), and social activities (M = 2.12). As was true of the national surveys of family violence, there is little difference in the amount of conflict between the ten issues argued about; for example the mean for the least frequent source of conflict (i.e.#10, pregnancy) was M = 1.46 whereas for the most common source of conflict (i.e. #1, managing money) it was M = 2.55, a difference of about one unit. Thus, conflict in these families was a fairly common occurrence.

Finally, a total conflict score was calculated for each case by summing the frequency of conflicts for each of the ten items. The possible range for the total conflict score was 10 to 50. For this sample, the mean of the total conflict score was 38.97 (SD = 9.18). This total conflict score also had a moderate correlation with whether the client was a battered woman (r = .60, p <.01).

To monitor PPRS agencies ACS collects on a monthly basis reports of caseworker activities related to each family. Included in this monthly report is whether the caseworker undertakes any activities related to domestic violence (e.g. counselling, referral, etc.). By examining the monthly reports for the year prior to implementing this project we were able to establish a baseline for the identification of domestic violence in the PPRS agencies which participated in the project. For the seven agencies participating in the first year of the project this domestic violence identification rate was 17%. The prevalence of domestic violence identified for the nine agencies added in the second year of the project was 15% This figure is a reflection of how well the caseworkers were able to detect domestic violence using non-systematic interviewing without the intake questionnaire.

One question in the Caseworker's Assessment section of the intake questionnaire asked whether domestic violence had been indicated on the family's original referral to the agency. In 92 (17%) of the 540 families the referral source had documented the presence of domestic violence in the family. Out of the 437 women who answered the initial question about domestic violence, 214 (48.9%) reported that they had experienced domestic violence in their current relationship.

The assessment questions on the extent of the violence indicated that many women were afraid of their partners and had experienced severe or even life-threatening abuse. In fact, 58 women reported that their partners physically assaulted them "regularly", 79 stated that it happened "sometimes" and 59 said it occurred "infrequently." The relationship between pregnancy and domestic violence documented by others (e.g. Walker, 1979) was also apparent in this sample, 102 women reported that there had been a fight during a pregnancy. Partners were also reported as trying to control the battered woman's activities regularly (n = 63), sometimes (n = 69), or infrequently (n = 21). While 179 women reported that their partner had threatened to hurt them, 76 stated that there partner had threatened to hurt himself and only 42 reported that the man had threatened to hurt their children. Threats to use a weapon were regularly experienced by 15 women and a sometimes or infrequent occurrence in the homes of 64 women. Seven women reported that their partners had regularly used a weapon against them and 11 stated that this sometimes occurred. For 31 women the use of a weapon was an infrequent occurrence.

Most of the women who reported that they had been battered also revealed that they had taken action in the past to deal with the domestic violence. One hundred seventy of the women had left their partner or had tried to leave because of domestic violence. The police had been called in 157 of the cases. Other actions such as obtaining an order of protection, counseling, going to a shelter, etc. were taken by 149 women. When directly asked by a caseworker at intake whether she wanted help at that moment to deal with the domestic violence, 65 women said yes.

Client Feedback

Initially caseworkers expressed concern that the domestic violence screening and assessment questions would be offensive or intrusive to some clients. The consumer satisfaction questions revealed almost universal acceptance toward being asked the domestic violence screening and assessment questions. About 2/3 of the women (n = 337) said that it was helpful to have been asked questions about domestic violence. Very few clients (n = 23) said that the questions could have been asked differently. Many of the clients (n = 134) said they felt better able to protect themselves or better able to protect their children (n = 121) as a result of having talked with their caseworker about domestic violence.

Battered Versus Not Battered Women

To explore what might differentiate battered women from women who were not battered in the population of these preventive service agencies socio-demographic, conflict and argument variables were analyzed using logistic regression. Logistic regression was utilized for several reasons; first the dependent variable (being a battered woman) was binary. Second, several of the independent variables could not meet the assumptions for discriminant analysis, most notably the assumptions of multivariate normality and equal co-variance matrices. Third, there seems to be a preference for logistic regression over discriminant analysis due to the additional information provided in the analysis (e.g. Dattalo, 1994; Morrow-Howell & Proctor, 1992). The purpose of this logistic analysis was to suggest inferences about the relationship between the independent (i.e. socio demographic variables) and dependent variable (being a battered woman) and to see how well the resulting model predicted group membership (Fraser, Jenson, Kiefer, & Popuang, 1994; Morrow Howell & Proctor, 1992)

Following a procedure recommended by Dattalo (1994), socio-demographic, conflict and argument variables were screened for inclusion into model development by examining their relationship to the group membership variable. Bi-variate t-tests and chi-square analyses which reached a liberal significant level of .15 led to a variable being included in model development. One reason for using this statistical criteria, in spite of the lack of any controls, is that the resulting model is much less likely to be overfitted (Dattalo, 1994). Furthermore, given the paucity of theory to explain the differences between battered women and non-battered women in the child welfare population, the use of this statistical selection process reduced the number of variables to be considered when building the final model. This procedure resulted in two continuous variables included in building the model: total conflict score (sum of all areas of conflict) and number of other people living in household. Four categorical variables also were statistically significant (p < .15): whether or not the woman's partner lived at home, whether or not at least one of the partners was employed outside the home, the race/ethnicity of the woman, and who had the last word in arguments.

Table 3 presents the results of a logistic regression model of the predictor variable of being a battered woman at intake in a Preventive Service Agency. The model correctly classified 71.7% of the women as either battered or not, 92% of those who were not battered were correctly classified and 74% of those who were battered were correctly classified. The model chi-square for this model was significant, 72 (2, N = 251) = 30.58, p < .001.

Table 3
Logistic regression model of being a battered woman at intake (1 = Battered Woman)

Variable		Coefficient	Standard Error	Odds Ratio
Total conflict		.06	.01**	1.07
Partner in home(1 = Yes)		-.79	.29**	.45
Constant			-.88	.51

*p < .05 **p < .01
(N = 251)

As table 3 shows, the odds of being a battered woman was 55% less likely if the woman's partner lived in the home. What this logistic regression suggests, on one hand, is that having a partner in the home decreases the odds of being a battered woman while on the other hand having more total conflict increases the odds of being battered.

To summarize, the results indicated that compliance with using the new intake questionnaire on all intakes was much greater in the first year (58%) when seven agencies participated in the project than in the second year (21.6%) when sixteen agencies were participating. Regardless of whether the woman was battered or not, these families experienced high levels of conflict. There was some relationship (r = .60) between total conflict and whether the woman was battered. The use of the questionnaire resulted in a much greater rate of identifying families where there was domestic violence, 15% to 17% without the questionnaire versus 48.9% with the questionnaire. It is also clear from the data that many of these battered women were victims of severe and life-threatening violence. Most of the battered women reported having taken action in the past to stop the abuse and many responded positively to the caseworker's offer of assistance. Finally, the majority of the women reported favorable reactions to having been asked questions about domestic violence.

While the rate of 48.9% for women identifying themselves as victims of domestic violence should be viewed cautiously, this rate is within the range for the percentages of cases reporting both domestic violence and child maltreatment in battered women's shelters (e.g. Bowker, Arvitell & McFerron, 1988; Gayford, 1975) and child protective services (e.g. Daro & Cohn, 1988; Hess, Folaron & Jefferson, 1992). The fact that almost three times more women reported being battered as had been referred with this noted underscores the need for comprehensive assessment upon intake in child welfare agencies. The high number of families who were experiencing or had experienced domestic violence is further proof of the pervasiveness of this problem.

One of the notable aspects of this project was the collection of data from the clients on their experience in being asked questions about domestic violence. We are unaware of any other reports where client feedback of this nature has been reported. The client feedback revealed women's universal acceptance towards being asked the question--they liked being asked. In Roddy Doyle's (1996) novel The Woman Who Walked Into Doors, about a battered woman in Ireland, the main character is quoted as saying to herself "Ask me, ask me, ask me," during a repeat visit to the local emergency room for bruises from a battering. Clearly this piece of fiction has deep echoes in the real world. "Ask me, ask me, ask me," may be the common refrain of women in private, confidential, personal interviews with a variety of helping professionals. And this study shows that not only do women want to be asked, but, they appreciate it.

A second significant finding from the consumer satisfaction data is that many women felt that having been asked the question about their being battered they felt better able to protect themselves, and their children. To what is this attributable? These are preventive service, neighborhood-based child welfare agencies. In a sense, as agencies not being identified as battered women's programs, and with the child as the identified point of intervention, women may have been more available for service. Many referrals for preventive services come from the child protection agency, with the balance often coming from local schools or health services. Women may have been more able to get services for themselves in the guise of getting services "mandated" for their children. Thus, this inadvertent strategy of child centered intervention may have lessened the batterer's suspicions of her therapeutic activities. And the fact that these services are neighborhood-based, and readily available many hours of the week, could only have aided her perception of their helpfulness. Whatever the reason, women felt better able to protect themselves and their children. This finding warrants both nurturance and further study.

The client feedback further underscores the importance of all workers being trained to routinely asking domestic violence screening and assessment questions. These client feedback data should assuage worker's fears about asking clients about domestic violence--the data demonstrate that asking is helpful to clients, their children, and to caseworkers in providing needed services.

The PPRS agencies who participated did so voluntarily--they had varying degrees of information and competence in dealing with domestic violence but an interest in the issue. Through training they expanded their competence and commitment to an issue oftentimes viewed outside the purview of traditional child welfare. The agencies' increased comfort and competence in dealing with this topic in the future is an unintended but significant achievement of the project.

It is laudable that in several of these agencies the increased identification of domestic violence lead to the development of new services for battered women. Several PPRS agencies, with the assistance of experts on domestic violence developed support groups for battered women. Another agency is working on developing a group for men who batter. Several workers continued their training and knowledge development about domestic violence in order to serve as agency-wide domestic violence specialists. Finally, four agencies, all located in the same borough, formed a domestic violence coordinating network.

These data also indicate that a large percentage of the battered women had sought help in the past, from a variety of sources, to deal with the violence they were experiencing. This suggests that these women are not "passive victims" or incapacitated by "learned helplessness" or "battered women's syndrome" but rather the help they obtain does not end the abuse. Our previous study of battered women identified during child protective service investigations found a similar pattern of prior help seeking behavior (Magen et al., 1995). An exploration of the reasons for the failure of services to end abuse are outside the realm of this paper, however we believe that these findings warrant further investigation.

There are several methodological and analytical limitations to this study. First, since the questionnaires were utilized with only 58% of new intakes in the first year and 21.6% in the second year, we do not know whether the remaining cases were systematically or randomly excluded. If there caseworkers made a purposeful effort to not utilize the questionnaire with certain cases then these data may be biased. In any event the results presented should be viewed in terms of their correlational nature and not as evidence of causal relationships. The logistic regression should be viewed cautiously given its exploratory rather than explanatory nature. Furthermore, there is a danger in interpreting the logistic regression in terms of a set of variables related to battered women and/or their environment which cause them to be battered. This project has operated under the assumption that battering is a form of control, with violence being the means to exercise that control. From this pro-feminist perspective, the cause of battering lies with the batterer and the social conditions that promote his behaviors. The value of the logistic regression is for caseworkers to identify which clients may be at relatively greater risk. Further research into the relationship between domestic violence and child maltreatment is needed for both practitioners and scholars.

Bowker, L. H., Arvitell, M., & McFerron, J. R. (1988). On the relationship between wife beating and child abuse. In. K. Yllo & M. Bograd (Eds). (1988). Feminist Perspectives on Wife Abuse (pp. 158-174). Newbury Park, CA. : Sage.

Centers for Disease Control (1995).

Citizens Committee for Children of New York (1997). Keeping track of New York City's children. New York: Author.

Cummings, N. & Mooney, A. (1988). Child protective workers and battered women's advocates: A strategy for family violence intervention. Response to the victimization of women and children, 11, 4-9.

Daro, D. & Cohn, A. H. (1988). Child maltreatment evaluation efforts: what have we learned? In G. T. Hotaling, D. Finkelhor, J. T. Kirkpatrick, & M. A. Straus (eds.) Coping with family violence: research and policy perspectives (pp. 275-287). Newbury Park, Ca.: Sage.

Dattalo, P. (1994). A comparison of discriminant analysis and logistic regression. Journal of Social Service Research, 19, 121-144.

Davis, J. (1995). Failure to protect and it's impact on battered mothers. Courts and Communities Confronting violence in the family, 1, 6-7.

Doyle, R. (1996). The Woman Who Walked Into Doors. New York: Viking.

Fraser, M. W., Jenson, J. M., Kiefer, D. & Popuang, C. (1994). Statistical methods for the analysis of critical life events. Social Work Research, 18, 163-177.

Gayford, J. (1975). Wife battering: a preliminary survey of 100 cases. British Medical Journal, 1, 194-197.

Hess, P. M., Folaron, G., & Jefferson, A. B. (1992). Effectiveness of family reunification services: an innovative evaluative model. Social Work, 37 (4), 304-311.

Magen, R., Conroy, K., Hess, P. M., Panciera, A., & Simon, B. L. (1995, July). Evaluation of a protocol to identify battered women during investigations of child abuse and neglect. Presentation at the 4th International Family Violence Research Conference, University of New Hampshire. Durham, New Hampshire.

Mahoney, M. R. (1991). Legal images of battered women: redefining the issue of separation. Michigan Law Review, 90, 1-94

McGowan, B. G. & Meezan, W. (Eds.) (1983). Child welfare: Current dilemmas, future directions. Itasca, IL: F.E. Peacock.

Morrow-Howell, N. & Proctor, E. K. (1992). The use of logistic regression in social work research. Journal of Social Service Research, 16, 87-104.

Schechter, S. (1982). Women and Male Violence. Boston, Ma.: South End Press.

Schechter, S. & Edleson, J. L. (1995, Spring). In the best interest of women and children: A call for collaboration between child welfare and domestic violence constituencies. The Prevention Report, 1-7.

Straus, M. A., Gelles, R. J., & Steinmetz, S. K. (1981). Behind closed doors: Violence in the American family. Garden City, New York: Anchor Press/Doubleday.

Straus, M. A. (1983). Ordinary violence, child abuse and wife beating: What do they have in common? In D. Finkelhor, R. J. Gelles, G. T. Hotaling, & M. A. Straus (Eds.), The Darkside of Families: Current Family Violence Research (pp. 213-234). Newbury Park, Ca.: Sage.

Stordeur, R. A. & Stille, R. (1989). Ending men's violence against their partners: One road to peace. Newbury Park, Ca.: Sage.

Walker, L. E. (1979). The Battered Woman. New York: Harper & Row.

Demographic Characteristics of Families (N= 540)

____________________________________________________________________________

Characteristic Woman Partner Children Others in Household

____________________________________________________________________________

N 540 328 1273 186

M n/a n/a 2.4 .4

SD n/a n/a 1.3 .9

Age

M 33.6 36.3 8.2 30.2

SD 8.5 10.8 4.7 20.1

Gender (N)

Male 0 320 674 74

Female 540 8 599 112

Source of Income (N)

Employment 129 172

Income Support 223 24

Social Security 17 3

SSI 37 15

Other 100 43 Table 2

Demographic Characteristics of Female Client (N= 540)

___________________________________________________________________________

Characteristic N

___________________________________________________________________________

Race

African-American 158

Asian 7

Caucasian 67

Hispanic 175

Native American 2

Other 12

Religion

Catholic 172

Jewish 10

Muslim 10

Protestant 50

Other 73 Table 3

Logistic regression model of being a battered woman at intake (1 = Battered Woman)

___________________________________________________________________________

Variable Coefficient Standard Error Odds Ratio

___________________________________________________________________________

Total conflict .06 .01** 1.07

Partner in home (1 = Yes) -.79 .29** .45

Constant -.88 .51

___________________________________________________________________________

*p < .05 **p < .01

(N = 251)

Instructions

This form must be completed for every female client whether or not you think there is violence. Interviews should take place in private, without partner or other family members present.

1) The domestic violence interview should take place at or within 30 days of intake.

A) If domestic violence is revealed at this time, fill out sections A (face sheet), B (interview), C (extent of domestic violence) and D (counselor's assessment).

B) If domestic violence is not revealed at this time, fill out sections A, B, and D.

C) Section E, the client assessment, should be completed after you and your client have had a chance to work together and not immediately after the domestic violence interview. Please wait at least 30 days (but not more than 90 days) before giving your client the assessment.

D) The case summary should be completed after the client has completed the client assessment.

Agency name: Counselor's initials: Code number:

Zip code of residency: Race: Religion:

Age Sex

Female Applicant

Child 1

Child 2

Child 3

Child 4

Child 5

Partner Relationship to applicant:

Does the partner reside in the home? Yes No

Others in household:

Sources of Income Female Applicant Partner

Employment

Income support

Social security

SSI

Other

Section B Interview

We are interested in learning more about your family, its needs and how we might provide you with better services. We hope that you will be willing to answer the following questions in as much detail as possible.

1. Who is in you home (both those who live there or visit frequently)? Enter numbers and relationships (e.g. 2 sons, husband lives there, aunt visits twice a week):

2. There are arguments in all families. The following is a list of things couples have told us they most frequently argue about. How often do you and your partner argue about these things? For each subject, place an "X" under the appropriate description of frequency.

Always Almost Always Usually Sometimes Never

Managing money

Cooking, cleaning, house work

Extended family/relatives

Social activities

Sexual relations

Discipline of the children

Things about the children

Drug or alcohol use

Other men/women

Pregnancy

Other (describe:

3. Who usually has the last word in an argument between you and your partner?

4. Has an argument ever turned into a physical fight between you and your partner?

5. Has there ever been a physical fight during a pregnancy?

6. In general, how afraid are you of your partner?

7. Domestic violence is a problem that affects many families in NYC of all different races, religions and incomes. Many women are abused by their intimate partners in emotional, sexual, physical, and materials ways. For example, some partners hit, push, punch, or kick. Others threaten them with harm, withhold money from them, force them to do things they don't want to do, or constantly criticize them. Do things like this happen in your relationship with your partner?

8. Has your partner ever threatened to hurt you?

9. Has your partner ever threatened to hurt your children?

10. Have you ever left your partner, or tried to leave, because of domestic violence?

Section C Extent of Domestic Violence

1. In general where are the children when you and your partner fight?

2. How do the children react when you and your partner fight?

3. Does your partner prevent you from leaving home, using the telephone, seeking family/friends, or otherwise control your activities?

4. Does your partner destroy your possessions or hurt things you value, including household pets?

5. Is your partner under the influence of drugs or alcohol when he abuses you?

6. Has your partner ever threatened to harm himself?

7. Has your partner ever assaulted you (punched, kicked, hit, pushed)?

8. Does your partner ever force you to have sex?

9. Does your partner ever force you to engage in sexual activities (including preventing you form having safe sex or using contraception)?

10. Has your partner ever threatened to use a weapon against you (gun, knife, hammer, or other object)?

11. Has your partner ever used a weapon against you?

12. Have you other others ever called the police because of domestic violence?

13. Have you ever gone to a doctor or hospital due to injuries caused by your partner?

14. Do you ever use drugs or alcohol as a way to cope with the abuse?

15. Have you ever asked for help to stop the abuse, such as police, a court order, counseling, support groups, shelter, family, friends, clergy or other outside help?

16. Would you like help with safety planning for you and your children?

17. Do you want help in seeking a temporary battered women's emergency residence, court order of protection, support group, or any other services for you and your children?

Section D Counselor's Assessment

1. Did this case come to you with domestic violence as part of the referral?

2. Was there domestic violence in a prior relationship or in the family history?

3. Is there a reason to suspect domestic violence in this family even though the client did not reveal this information in the interview?

4. If domestic violence is a current problem, have you tried any of the following resources and what were the results?

Shelter

Hotline

Doctor or Hospital

Police

Legal Services

Court

Support Group

Counseling

Other

Section E Client Assessment

1. Do you feel that it was helpful to be asked questions about domestic violence?

2. Was there a way we could have asked the questions that would have been more helpful?

3. Was any of the information provided to you about domestic violence helpful to you or anyone else?

4. Do you feel that you have been able to better protect yourself because of conversations about domestic violence with your counselor?

5. Do you feel that you have been able to better protect your children because of conversations about domestic violence with your counselor?

6. How can we be more helpful in dealing with domestic violence in your family?