‘I .0 afiawfiérz (”tic . Sf ;,.fw.,rt f, 5: 1...... . ‘21:}; 1:41‘ :1... ‘ L .. t “qfii' :1 u "l 7 o... «.3 I... 2 '9 .. .1 J. ‘ b. . 3. .. Ii. daflfl.3‘ i ”1....mm1l3 1 H531". nu; figs. hm: mm£§§3 gunfigym? E M313 0 1111111111111111111 3 1293 01719 7343 Michigan State University This is to certify that the thesis entitled EVALUATION OF A PARENT EDUCATION PROGRAM FOR LOW RISK MOTHERS , presented by Camilla Ruth Williams has been accepted towards fulfillment of the requirements for Haster of Artdsegree in Psychology Major professor Date [y/‘I/‘f! 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution DATE DUE DATE DUE DATE DUE .. . mm")? Win '7 "31-. FEB 20' 2000 a 430mb? tot, t1! ”TOE.“ “ ‘ .fllzoon PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. MAY BE RECAU£D with earlier due date if requested. 1“ WM“ EVALUATION OF A PARENT EDUCATION PROGRAM FOR LOW RISK MOTHERS By Camilla Ruth Williams A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Psychology 1998 ABSTRACT EVALUATION OF A PARENT EDUCATION PROGRAM FOR LOW RISK MOTHERS By Camilla Ruth Williams Program planners, service providers, and evaluators are searching for the best practices in preventing child abuse. In this study, an evaluation was made of a child abuse prevention program for low risk mothers. The goal of the evaluation was to examine the roles of program content, program intensity, and parent characteristics on program effectiveness. However, no differences were found between participants and community comparisons. Possible explanations for the lack of results are discussed. ACKNOWLEDGMENTS Writing this thesis has been a long and often difficult experience for me. I would like to recognize the people who have supported me throughout the process. First of all, I would like to thank my committee members. Bob Caldwell, my advisor and thesis chair, has encouraged me to put my thoughts on paper and helped me organize them. Anne Bogat has reminded me that what seems implicitly obvious needs to be explicitly communicated. Tom Reischl has guided me into new ways of thinking about clinical and community psychology. With their guidance, 1 am learning about my profession and about myself. I would also like to thank two people who have kept me going through the detours and seeming dead-ends. My husband, Carrick, has supported me both instrumentally and emotionally. As my statistics consultant and editor, he endured many late night brainstorming and critical thinking sessions. More importantly, he is a dear friend who knows when to problem solve, when I need a little push, and when to listen quietly. Jodi Levy-Cushman has been a friend with whom I can share both the dreams and the frustrations of our chosen profession. Her faith in me has renewed my faith in myself, and her strength has supported me when I wasn’t sure I could support myself. iii TABLE OF CONTENTS LIST OF TABLES ...................................................... vi INTRODUCTION ....................................................... 1 Prevention Through Parent Education and Family Support Programs ......... 3 Prevention Program Characteristics .................................... 5 Program Content ............................................ 6 Developmental Information .............................. 7 Social Isolation and Social Support ........................ 7 Community Resources .................................. 9 Program Length ............................................. 9 Target Population ........................................... 10 Program Format ............................................ 12 Proactivity Versus Responsiveness ....................... 12 Qualifications of Service Providers ....................... l4 PROVIDING PARENT EDUCATION AND SUPPORT: A CASE STUDY ........ 15 Hypotheses ...................................................... 16 Method ......................................................... 17 Program Characteristics ...................................... 17 Parenting Information Packet ........................... l8 Volunteer Characteristics ............................... 18 Telephone Calls ...................................... 19 Participants ................................................ 20 Measures ................................................. 25 Effects of Program Length on Community Resource Knowledge 26 Measures of community resources knowledge ........ 26 Effects of Program Length on Support .................... 28 Number of support people mentioned ............... 28 Adequacy of resources ........................... 28 Effects of Program Intensity on Participant Benefit .......... 29 Number of calls received ......................... 29 Role of informational packets. ..................... 29 Role of phone calls .............................. 29 Effects of Parenting Experience on Participant Benefit ........ 29 Parenting experience ............................ 30 Benefits of the program: support and information ...... 30 Results ......................................................... 31 Discussion ...................................................... 34 iv APPENDICES ......................................................... 39 Appendix A: Time 1 Intervention Group Protocol ....................... 40 Appendix B: Time 1 Comparison Group Protocol ....................... 48 Appendix C: Time 2 Interview Protocol ............................... 55 Appendix D: Time 2 Intervention Interview Consent Form and Cover Sheet . . 63 Appendix E Time 2 Comparison Interview Consent Form and Cover Sheet . . . 66 LIST OF REFERENCES ................................................. 69 LIST OF TABLES Table 1: Hospital Referral Form Risk Factors Identified ................................................... 22 Table 2: Demographic Information .................................................................................... 23 Table 3: Group Size, Mean, and Standard Deviations for Analyses .................................. 32 vi INTRODUCTION Child abuse is a traumatic experience that creates untold immediate and long-term suffering for those who experience it. Immediate effects of abuse include the following: increased anger and aggression, interpersonal difficulties, decreased self-esteem, dependency, depression, dissociation, and academic problems (Briere, 1988; Reppucci, Britner, & Woolard, 1997). Long-term effects of abuse on mother-child relationships have been studied and show that mothers who were abused as children may have difficulty interacting with their children. This difficulty bonding with their own children perpetuates a cycle of poor parent-child relationships across generations of a family (Olds, 1988; Whipple & Wilson, 1996). Associations have also been made between child maltreatment and juvenile delinquency and later violent and criminal behavior (Briere, 1988). Those who have been abused also tend to have increased psychological symptoms, including low self-esteem, and they are more likely to abuse their own children than parents who were not abused as children (Briere, 1988). One way to combat the problem of child abuse is to offer treatment for families that have been identified as abusive. Common services that are required to deal with child abuse and neglect include hospitalizations, rehabilitation, foster care, special education services, social service case management, court expenses, and treatment programs for parents and children (Westrnan, 1994). However, these post-abuse services require a substantial amount of financial resources (Westman, 1994), and, in spite of early, well-planned, and costly post-abuse intervention efforts, child abuse has been shown to continue to occur during treatment (Cohn, 1979; Cohn & Daro, 1987). Because of the high financial and emotional costs involved in treating child abuse, and the limited effectiveness of this treatment, communities are looking to prevention as an alternative approach for decreasing the incidence of child abuse (MacMillan, MacMillan, Offord, Griffith, & MacMillan, 1994). Child abuse prevention and post-abuse intervention programs target the same problems. However, there are many advantages of prevention over treatment. One critical advantage is that prevention reaches its population before violence occurs. By reducing the incidence of child abuse, prevention programs also decrease the human suffering that results from abuse (Reppucci et al., 1997). Additionally, prevention programs are able to use limited resources such as time, money, and expertise more efficiently than treatment programs (Harper & Balch, 1975). There are two approaches to prevention that conceptualize childrearing and child abuse in different ways: primary and secondary prevention. Primary prevention is a universal approach to prevention that attempts to prevent abuse by enhancing parenting skills for all parents. Secondary prevention, on the other hand, attempts to prevent abuse from occurring in families that have already been identified as being at risk for abuse. Advocates of secondary prevention (e.g., Pillow, Sandler, Braver, Wolchik, and Gerstan, 1991) suggest that targeting families that have an increased likelihood of becoming abusive allows programs to maximize their effectiveness by focusing only on those who need the services the most. However, Caldwell (1991) cautions that screening potential service recipients has several drawbacks and may be more costly than providing the services to everyone. This is true not only in terms of financial expenses, but also because of the potential cost in human suffering for mislabeling families (both false positives and false negatives). One benefit of the primary prevention approach is that people may be more likely to respond to its universal nature because services target "normal" pe0ple, not those who have been identified as having problems. This lack of stigma may encourage parents to be involved in a primary child abuse prevention program, whereas becoming involved in secondary prevention services may be seen as an admission of deficits in parenting abilities. Guterrnan (1997) points out that the universal approach to prevention programs may have a greater impact on the community because participation is more fully voluntary, whereas the risk screening process may eliminate parents who are receptive to prevention services. Therefore, primary prevention programs make themselves accessible to a broader range of individuals than programs limiting themselves to high risk individuals (Guterman, 1997). Prevention Through Parent Education and Family Support Programs The goal of primary prevention of child abuse is to reach healthy families that may or may not be at risk of abuse and to strengthen their resources and abilities to handle the rigors of childrearing before a problem exists (Rosenberg & Reppucci, 1985). According to this approach, everyone will encounter stress at some point in their lives, but they will handle that stress in different ways. For parents, the birth of a child has been highlighted as a particularly stressful time, especially for first-time (primiparous) parents (Guterman, 1997). This stress results from increased physical and emotional demands, as well as unexpected changes in the marital relationship. Along with these changes comes the recognition of the restriction in roles and opportunities and a shift in personal support networks, especially for young mothers (Reppucci et al., 1997). Under the stressful conditions of learning a new role and attempting to deve10p a relationship with a new child, some parents will suffer emotional difficulties and may need outside support. However, other parents may grow into the new role and enj oy the challenges, while still others will not appear to be either positively or negatively affected by becoming parents. Because we cannot predict how a given person will react to the birth of his or her child, it is difficult to know who will need and benefit most from child abuse prevention services (Caldwell, 1991). Therefore, primary prevention programs focus on enhancing existing parenting capabilities, increasing available resources, and highlighting parents’ current coping abilities under the assumption that all parents can benefit from these services (Rosenberg & Reppucci, 1985). These goals can be accomplished through parent education, family support, or a combination of both education and support. Parent education programs assume that all parents want to do what is best for their children, but some parents lack knowledge of and experience with childrearing techniques. Therefore, supporters of the educational approach believe that teaching appropriate child development and age-appropriate discipline will modify inappropriate parenting practices and decrease the stress in parents’ lives, which will decrease the likelihood of abuse (Reppucci et al., 1997). This belief is supported by Guterman’s (1997) review of 18 primary prevention programs, in which he noted that all programs that reported positive outcomes incorporated some form of parent education. Family support programs believe that all families have strengths and weaknesses that affect the way the family functions within the community. The goal of providing support is to increase the family’s ability to cope with problems by building on family strengths. In order to foster these strengths and reduce weaknesses, the family support model believes that it is essential for the family to have a social support network. Because the family is part of the community, its needs cannot be met in isolation from the community. Therefore, support should be provided in the context of community life through links to appropriate community resources. Guterman (1997) cites evidence from several studies supporting the notion that the effort to link families to necessary formal and informal supports is a core ingredient of successful child abuse prevention. Although parent education and family support are separate approaches to primary prevention, they are often both incorporated into prevention programs. It has been argued that parent education programming must occur simultaneously with family support efforts and attempts to decrease situational problems that interfere with child-rearing in order to be effective (Reppucci et al., 1997). Proponents of the joint education and support model point out that a parent is less likely to benefit fi'om information about enhancement or health issues when he or she does not feel secure that basic family needs are being met. Prevention Program Characteristics Prevention programs have attempted to decrease the incidence of abuse using a variety of program models, and results indicate that not all programs are equally effective for all families (Guterman 1997; Reppucci et al., 1997). When considering the most effective way to reach parents, target family variables such as demographics, social network ties, and input or involvement levels in the program must be taken into account (Reppucci et al., 1997). Additionally, program effectiveness is affected by program factors such as the level of training and commitment of volunteers, the nature of the parent-volunteer relationship, program length, and setting (Reppucci et al., 1997). It has been noted that the most successful programs have begun to change fiom standardized, curriculum-based classes to individually-tailored programs that are culturally responsive and contextually relevant to the family (Reppucci et al., 1997). However, program planners and researchers still debate basic programming issues such as what content should be included in programs (e.g., community resources, social support, child development, etc.), how long programs should last (i. e., short-term or long-term), what population should be served (i. e., first-time parents or parents with other children), and how services should be provided (i. e., through home visits or in some other venue). mm One of the first decisions to be made when planning prevention programming is what kind of information will be provided to program participants. Child abuse prevention programs have included a broad range of services ranging from infant first aid to self-care for parents and marital or family counseling. Three broad educational content areas that are often incorporated into prevention programs are developmental information, social support, and community resources. Developmental Information Education about child development is an important content area for primary prevention programs. Parents often do not set age-appropriate standards for their child’s behavior because they do not understand the abilities and limitations that childhood brings (MacMillan et al., 1994). Resnick (1985) found that parent training appears to 1 improve the quality of the parent-child relationship, at least in the short term. Additionally, Taylor and Beauchamp (1988) found in their work with a hospital-based primary prevention program that new parents who received developmental information had more realistic expectations for their children than those who did not receive this information. Inappropriate expectations are thought to lead to parental frustration and conflict, and children may be seen as obstinate in situations where they are not able to meet parent expectations (Resnick, 1985; Taylor & Beauchamp, 1988). Social Isolation and Social Supmrt Another factor that has been highly correlated with abuse, and is therefore targeted by prevention programs, is social isolation (Olds, 1988; Salzinger, Kaplan, & Artemyeff, 1983). Socially isolated mothers tend to have fewer members in their social network, and sources of support tend not to be in contact with one another, which means that coordination of support is difficult (Salzinger, Kaplan, & Artemyeff, 1983). A link has also been made between the types of support people on one’s network and abuse likelihood. Specifically, Salzinger et al. (1983) conducted a network analysis of the support systems of abusive mothers that showed a tendency toward a small number of family members with few or no supports outside of the family circle. Often, these support networks are causes of stress as well as support, and the nature of the kinship bond is such that corrective feedback may not be provided for inappropriate parenting behaviors. In fact, families ofien transmit and perpetuate unsuccessful parenting styles and abusive patterns across generations (Salzinger et al., 1983). Decreasing social isolation and enhancing social support for parents (especially mothers) is an important aspect of child abuse prevention programs. According to Krugman (1995), the best method for preventing child abuse and neglect appears to be linking new parents with a supportive friend, an extended family member, or a professional support provider such as a public health nurse. Contact with people in the community decreases social isolation, and linking parents with appropriate community resources reduces the stress that can lead to parental depression (Olds, 1988). Contacts with community resources that provide support will encourage the mother to feel less lonely, and the feedback about appropriate and inappropriate parenting behaviors can provide incentives for a mother to parent more effectively as well as reinforcing her sense of competence as a parent (Salzinger et al., 1983). Of note, an increased sense of control in parenting situations has been linked with less maltreatment and fewer visits to the hospital emergency room (Olds, 1988). In addition, when parents are successfully involved in educational and support services, competence in reaching out to others will increase and confidence in ability to solve problems will encourage more independence (Taylor & Beauchamp, 1988). The support and reinforcement of parenting competencies will lead parents to feel better about themselves as parents. Communig Resources Although researchers and program planners value support and education services as critical for prevention programs, parents are often more interested in being provided with concrete services rather than abstract support (Barth & Ash, 1986). Therefore, parents may not value services that are limited to decreasing social isolation. In order to accommodate family needs and desires, prevention programs that provide educational and support services often incorporate information about community resources (e.g., food and clothing banks, financial assistance, or child enrichment programs). Participation in parent education programs also increases knowledge about community resources. For example, Taylor and Beauchamp (1988) found that at three months postpartum, a comparison of participating and non-participating mothers showed that non-participating mothers had fewer solutions to problems that arose in child rearing, and that they were able to cite fewer sources of assistance in dealing with these problems. These authors state that the enhanced problem-solving capabilities gained through participation in educational programs are maintained into later childhood, but they caution that longitudinal evaluation of this claim is necessary. Pro Len The question of length of parent education programs has rarely been addressed in the literature, and the results thus far are mixed. Taylor and Beauchamp (1988) found indicators of positive effects of education, including increased capability to handle parenting crises, that were maintained over a three month post-treatrnent time period. Weinman, Schreiber, and Robinson (1992) also found increases in parenting knowledge from pre-treatrnent levels to post-treatment levels in a two month parent education program for adolescent mothers. These increases were maintained at eight weeks post- treatment, except for the measure of inappropriate developmental expectations. Whipple and Wilson (1996) also emphasize that it may be possible to change some types of parental attitudes and increase knowledge about parenting in a short amount of time. However, they suggest that long-term participation in education programs may be required in order to consolidate these initial gains. Guterman (1997) cautions that short-term gains measured soon after program completion may fade over time. He notes in a comparison of short-term programs that those reporting success relied on follow-up studies occurring soon after program completion, while those not reporting success utilized more long-term follow-up measures. In particular, Guterman (1997) pointed to a longitudinal study utilizing multiple measurement times in which early positive intervention effects had disappeared at eight months and beyond. W Parent education programs often target primiparous parents for primary f prevention programs. It is thought that these parents, who are making the transition from adult or couple into parent and family status, are at an especially vulnerable time in their lives (Zigler & Black, 1989). Primiparous parents are thought to be in need of a broader range of services because of the novelty of the parenting role and because they are less likely than experienced (multiparous) parents to have established networks to access when parenting problems arise (Whipple & Wilson, 1996). The fact that they are in 10 transition also suggests that they may be more willing to accept information and advice than multiparous parents are, and therefore, they are at a peak moment to receive prevention information (Guterman, 1997; Taylor & Beauchamp, 1988). Olds (1988) also suggests that targeting primiparous parents is the most cost-effective way to conduct prevention because skills and resources developed when caring for the first child will be maintained with subsequent children. It is believed that in order to have long-lasting effects, programs need to reach families at the earliest moments in the child’s life as well as the earliest moments in the parent’s new role (Guterman, 1997). Although some theorists suggest focusing prevention efforts on primiparous parents, many of the arguments made for targeting new parents also apply to multiparous parents. For example, Guterman (1997) states that the initial phases of the first parent- child relationship are the most vulnerable and provide the greatest opportunity for intervention to establish positive interaction patterns. However, Goldberg and Michaels (1988) point out that this is true for all children in a family, not only the first-bom. Because each child is unique, his or her interactions with the parent will differ from those of siblings. Thus, it is important to teach multiparous parents how to respond to the individual needs of the new child. In conjunction with the uniqueness of each child’s temperament, Goldberg and Michaels (1988) point out that each "transition-to-parenthood" experience is unique. Parents who were well-supported during their first pregnancy may feel abandoned when subsequent pregnancies are not met with equal support. Other experiences, such as 11 integrating the new child with the existing family system, are stressful and may increase the risk of family violence (Goldberg & Michaels, 1988). Prom Format A final important consideration for program planners is the mode of service provision. Some programs reach out to possible clients, while other programs place the responsibility onto clients to seek help. Programs may be offered at hospitals or local community centers, or they may work in the participants’ homes. Some programs employ a staff of professionals such as public health nurses or social workers, while others rely on paraprofessionals or volunteers. Progtivity Versus Resmnsiveness Traditionally, social service programs have been based in a central location and have relied upon individuals to seek them out. However, Gottlieb and Pancer (1988) suggest that new parents may have difficulty obtaining the parenting support they need because of the demands of parenting. These authors state that new parents are often housebound and may be too fatigued from childcare pressures to pursue outside resources. Parents may also be reluctant to seek out childcare assistance because they fear that they will be seen as inadequate parents (Gottlieb & Pancer, 1988; Guterman, 1997). An alternative to center-based services is a home visiting program in which the service providers come into the family home to provide support. Several studies have demonstrated the effectiveness of home visiting programs in preventing child abuse with high-risk parents. For example, Olds (1988) evaluated an intensive home visiting 12 program with primiparous mothers who were rated as high-risk because they were classified as low socioeconomic status and were unmarried teenaged women. Additionally, the Hawaii Healthy Start Program, which is ofien cited as a model of the best prevention practices available, utilizes home visitors to decrease the likelihood of abuse for high-risk mothers (Guterman, 1997; Wallach & Lister, 1995). Although evidence exists that home-visiting programs are effective in preventing child abuse, neither of the programs mentioned above offered services for mothers at lower risk. In fact, the evaluation of programs for low-risk parents is lacking. One reason for the lack of evaluations of programs for low-risk parents is that these programs are rare. Home visiting programs are costly to maintain, and thus service providers have concentrated their attention on high-risk parents because they are perceived to need services and benefit from them more than low-risk parents do. Pillow and colleagues (1991) argue that lower risk parents have less to gain from prevention services and therefore will not show improvement as a result of services that have been provided to them. Furthermore, for child abuse prevention programs that serve all families, some researchers attribute the failure of these programs to the fact that they did not selectively target only the highest risk mothers (Siegel, Bauman, Schaefer, Saunders, & Ingram, 1980). Until less expensive programs are available, service delivery providers will continue to limit services to the families with the highest risk in order to see the maximum return on their investments. This logic of only serving the neediest people in order to show that programs produce results confounds evaluation research of the best 13 practices for promoting healthy families. The question remains as to the effectiveness of prevention programs, especially for low-risk mothers. Qu_alifications of Service Providers The debate also continues about who is qualified to provide services to at-risk families. Traditional treatment and educational services have been provided by professionals working within formal human services organizations (Gottlieb & Pancer, 1988). More recently, as program funding becomes more difficult to obtain and as programs move into the community, responsibility for service provision has shifted to volunteers (Gottlieb & Pancer, 1988; Reppucci et al., 1997). Reppucci and his colleagues (1997) highlight the benefits of utilizing volunteers to provide services. They state that volunteers provide free assistance, foster community ownership, and provide outstanding services. However, volunteers also require administrative management, recruitment and retention efforts, and training. Therefore, prevention programs that rely on volunteers for service provision must still provide staff oversight of volunteer activities. For primary prevention programs, Martin, Scott, Pierron, and Bauerle (1984) suggest that more effective prevention can be accomplished in the long-term by community-based organizations utilizing volunteers than is provided by large agencies with a large infrastructure to support. This suggestion is supported by research suggesting that volunteers or (para)professionals who are parents themselves may be able to connect better with parent reactions and can share personal life experience and wisdom (Olds, 1988). However, program creators must work to find the balance between 14 creating rapport between the volunteers and participant families and fostering a relationship where program participants become dependent on volunteers (Olds, 1988). PROVIDING PARENT EDUCATION AND SUPPORT: A CASE STUDY Although theorists agree that parent education and family support are important aspects of successful prevention programs, evaluations of specific programs suggest that continuing research is needed to find the best prevention strategies. In particular, the following questions about child abuse prevention remain: a) It appears that not all prevention programs are equally effective for all families. What factors of a program (e.g., program length and content) and a participant family (e.g., primiparous or multiparous parents) promote or inhibit success? b) Is primary prevention effective for parents identified as being at low risk for abuse?, and c) how long must prevention efforts continue in order for them to be successful? Is a single exposure "inoculation" of prevention information enough, or do "booster shots" need to be administered? The purpose of this study is to evaluate the importance of program length and program content in the success of parent education programs. In particular, the Kent County Healthy Start (KCHS) parent education program will serve as a case study of a child abuse prevention program that offers services to parents identified as being at low risk for child abuse. The program goals are to provide social support to these parents, as well as to provide them with information about their child’s developmental growth and community resources that are available to address parenting concerns. 15 Hypotheses The comparison groups in this study consist of mothers who participated in an 8 week program, mothers who participated in a 12 month program, and comparison mothers who did not participate in either program. All participants (8 week, 12 month, and comparison participants) were contacted for interviews at two times so that two sets of data were collected: 8 week data and 12 month data. 1. Participants in the 12 month program, as measured at the end of 12 months, will benefit from the KCHS program more than those in the 8 week program, as measured at 8 weeks. Both of these groups will differ from the comparable nonintervention comparison group. Specific differences include the following: a. Informational differences: Parents who participate in the 12 month program will be more knowledgeable about appropriate community resources than those in the 8 week program, and both groups of participants will be more knowledgeable than respective comparison samples. b. Support differences: Twelve month participants will mention more support people than 8 week participants, and both groups will mention more support people than respective comparison samples. In addition, a greater proportion of participant than comparison mothers will state that they feel that they have enough help to address parenting concerns. We anticipate the same pattern of results, namely that more 12 month participants will endorse that statement than 8 week participants. 16 2. For all participants, the more calls a participant received, the more likely she is to say in an interview at 12 months that receiving the informational packet and receiving calls from volunteers helped her to become a better parent. 3. Previous parenting experiences will influence the ways in which participants gain from an education intervention. These experiential differences will lead to the following response patterns to the 12 month interview: a. Primiparous mothers are in need of a broader range of services. Therefore, a higher proportion of primiparous than multiparous mothers will list information as one of the most helpful aspects of the intervention phone calls. b. Because they already have a support network in place, a greater proportion of multiparous than primiparous mothers will either not state that anything in the program was helpful or will have negative comments about the program’s benefits. Method mm Characteristics Kent County Healthy Start (KCHS) is a child abuse prevention program operating in Kent County, Michigan, that targets women who give birth in hospitals in Grand Rapids, Michigan. KCHS offers a home visiting program for women who are considered to be at high risk for abuse and a parent education program for women who are considered to be at low risk. This study will focus on the parent education portion of the project. 17 The parent education intervention was created as a low cost prevention strategy to support new mothers during a stressful time in their lives. The program goals are to promote positive relationships between parents and their children, to provide information about a child’s developmental growth and needs, to make parents aware of community resources that are available to address parenting concerns, and to provide social support to parents. These goals are accomplished by mailing a comprehensive packet of information to families two weeks after the birth of their babies and following this information with supportive phone calls provided by a volunteer. Parentjpg Information Packet Parents were contacted about participating in the KCHS parent education program while they were in the hospital. Eligible mothers that chose to participate received a packet of informational materials about community resources in the mail approximately two weeks after giving birth. This packet included pamphlets about infant care and feeding, immunization, infant safety, child development, encouraging literacy, child care, and family support services. In addition, families were notified of the volunteer telephone schedule and were given the phone number to contact KCHS if they had any questions. Volunteer Characteristics The KCHS volunteers who made telephone calls to project participants were successful mothers recruited from the Grand Rapids community. These women were trained in telephone protocol and community resources, and they attended monthly training sessions to update their knowledge of the community and to share insights and 18 questions about the program. The program was designed so that a single volunteer would contact families throughout the length of the program. However, if participant mothers returned to work or changed their schedules and needed to be contacted at a different time, a different volunteer might become the new support person. Changes in volunteer contacts due to changes in the mother’s support needs or to volunteer turnover were not tracked by the program. Telephone fils The role of the volunteer was to act as a support person for the mother in her parenting role and to discuss life changes that occur with a newborn in the house. This role included making sure that the family received the original information packet, discovering how the family was adjusting to the baby, and connecting the family to appropriate resources when needed. The KCHS Volunteer Coordinator created a list of conversation starters and potential questions to ask families, and volunteers kept a written record of parent issues and concerns that arose during phone calls. This record was used to track continuing family needs and family progress. Although the program was designed so that a standard number of calls were made at evenly spaced intervals, families that experienced problems were contacted more frequently in order to provide ongoing support. Additionally, mothers who were not interested in continuing contact with KCHS could request that they not receive any more phone calls. Originally, the phone calls were designed to be made approximately four weeks and eight weeks after the child’s birth, but this design was modified to a twelve month program within the first year of the program’s existence in order to address continuing l9 parenting concerns. For the six month transition period between the 8 week and 12 month programs, participants were given the option of continued contact with Healthy Start volunteers beyond the eight weeks. For those who wished to continue, additional calls were made at months three, six, nine, and twelve in order to target critical developmental periods. The alteration in the design of the program created two intervention groups that could be compared, namely those who participated in the program for eight weeks and those who continued to participate for twelve months. In addition, a group of women who were assessed in the hospital and met the criteria for inclusion in the parent education program but were not offered the option to participate formed a comparison group. Participants The two hospitals that participated in the KCHS parent education program served residents of Kent County as well as those who lived outside of the county. For the purposes of the Healthy Start program, however, only Kent County residents were eligible to receive services. According to Butterworth Hospital statistics, there were approximately 5700 total births at the hospital in 1997, and 4100 (72%) of those births were to Kent County residents. St. Mary’s Hospital statistics indicated approximately 2,000 births, 1200 (60%) of them to Kent County residents. Roughly 30% of the 5300 families that were eligible for the KCHS program were not contacted because the births occurred during the weekend, which brought the potential number of participants to approximately 3700 mothers. Other families were not considered for the parent 20 education program because they were considered to be at higher risk for child abuse. Of the mothers who satisfied the criteria for participating in the parent education program, approximately 60% were offered KCHS services, while the other 40% were approached to serve as comparison mothers for the purposes of evaluation (based on the day of the week on which they gave birth). Overall, 232 families participated in the KCHS parent education program in 1997, and 75 of these mothers were involved in the program evaluation. The birth rate information from both hospitals for the year 1996 is similar to the 1997 numbers. During 1996, 65 of the 229 mothers that participated in the parent education program were involved in the evaluation process. A general hospital referral form was used in the participating hospitals to assess participants. This form included a variety of health risk factors, some of which also have been identified as risk factors for child abuse. A part-time hospital employee affiliated with the hospital social work department filled out the forms based on hospital charts and records. Women with fewer than two risk factors were offered the opportunity to participate in the parent education program, while women with two or more risk factors were offered the opportunity to participate in the home visiting program. Table 1 identifies the risk factors from the referral form that were endorsed by at least one participant in the parent education program. 21 Table 1 Hospital Refer_'r_aJ Form Risk Factors Identified Risk Factors Program Phases 8 week 12 month Comparison Overall None 2 Jaundice Meconium Aspiration Congenital Anomalies 5 3 0 1 1 Feeding Problems 1 1 0 3 4 3 p—n O N Premature Unplanned/unwanted Pregnancy Low Functioning or Mental Illness Pre-existing Condition (specified 2 in chart notes) Single Parent Reproductive History Severe Prenatal Complications Severe Perinatal Complications Overweight/underweight Smoking Financial Need Unemployed Substance Abuse Parenting Concerns Isolation or Lack of Support Medical/genetic History Total ~o~oo~ooxo coo—ooNu—oo DJ Am--u~ p—s —o N mw—N~Nw~—~oo— l l 208 OOUINOH—‘OOOHUJH t--'t-“C>'--H'=*OOF‘W-‘OO-hCD \rocoo—o—Nr—o—o 00 Kit \)r—I In addition to filling out the hospital referral form, this same part-time hospital employee interviewed potential program participants in order to assure that chart information was correct and to discuss any parenting concerns. These interviews were conducted with mothers who gave birth on weekdays, and the eligible women were placed in either the participant group or the comparison group depending on the day the interview was done (for demographic information, see Table 2). Mothers who gave birth 22 on weekends or who left the hospital before an interview could be conducted were not included in the program. Table 2 Demographic Information Demographic Information Program Phases 8 week 12 month Comparison Overall Completed Interviews 80 51 77 208 Primiparous 44 13 20 77 Multiparous 34 36 57 127 Missing Data 2 2 0 4 Age M (SD) 29.0 (5.2) 29.61 (4.22) 30.05 (4.16) 29.66 (4.4) Range 20-45 22-43 23-39 20-45 Previous Pregnancy M (SD) 1.06 (0.9) 0.64 (0.87) 1.22 (0.93) 0.96 (0.93) Range 0-3 0-3 0-3 0-3 23 There were 80 women who participated in the eight week program. These women ranged in age from 22-43 years (mean = 29.61, standard deviation = 4.22), and ranged from 0 to 3 previous pregnancies (mean = 0.64, standard deviation = 0.87). The group that participated for 12 months was comprised of two subgroups: women who participated during the transition between the eight week and twelve month programs and could choose whether or not they wanted to continue in the program after eight weeks, and women who participated in the twelve month program after the transition period and therefore were not given the choice of leaving the program afier eight weeks. In the choice group, the 29 women ranged in age from 20-45 years (mean = 29.26, standard deviation = 5.55), and ranged from 0 to 3 previous pregnancies (mean = 0.85, standard deviation = 1.01). The 22 women in the no-choice group ranged in age from 21-36 years (mean = 28.62, standard deviation = 4.80), and ranged from 1 to 3 previous pregnancies (mean = 1.32, standard deviation = 0.65). Two t-tests conducted on this subsample of women showed that they did not differ in age (1 (37) = 0.37, p = 0.71) or in the number of previous pregnancies (t (47) = -l .85, p = 0.07). A chi-square analysis conducted to determine if the choice and no-choice participants differed in number of risk factors indicated that the two groups did not differ (x2 < 0.01 , p > 0.05). Therefore, they will be considered as one group. For this group of 51 women who participated in the twelve month program, the age ranged from 20-45 years (mean = 29.00, standard deviation = 5.20), and ranged from 0 to 3 previous pregnancies (mean = 1.06, standard deviation = 0.90). 24 The comparison group consisted of 77 women. These women ranged in age from 23-39 years (mean = 30.05, standard deviation = 4.16). They had a previous pregnancy range of 0-3 (mean = 1.22, standard deviation = 0.93). Two t-tests were conducted to determine whether these women differed from the combined participant sample on demographic variables. There was no significant age difference between participants and comparisons (1(193) = -0.99, p = 0.32). Although the difference between groups in number of children was significant (1 (202) = -3.18, p < .01), this difference was less than one child. Measures Two interview instruments were created to assess the participants’ satisfaction with the program. Specifically, we inquired about their satisfaction with the information they received about child development and community resources. We also asked about their sense of being supported by the KCHS volunteer. Both of these interviews were designed to be conducted over the telephone, the first approximately three months afier entry into the program (see Appendix A for intervention group protocol and Appendix B for comparison group protocol) and the second approximately thirteen months after entry into the program (see Appendix C for the measure, Appendix D for the intervention consent form and coversheet, and Appendix E for the comparison consent form and coversheet). Eight week participants, twelve month participants, and non-participant comparisons all received telephone interviews at both times. KCHS volunteers conducted interviews with program participants, while two local data collectors were 25 hired to conduct interviews with the comparison group members. The following variables were taken from the interview data. Em o_f m Lgrgm pp Knowledge o_f Community Resources Hypothesis 1a refers to the amount of benefit that a participant gained by being involved in the extended 12 month parent education program rather than the 8 week program or not being involved at all. It states that the longer a participant is involved in the program, the greater her knowledge of community resources will be. Four questions on the interview protocol are related to this hypothesis, and each was considered in turn before the final operational definition was made. Measures if communig resource knowledge. In both of the interviews described above, mothers were asked about their knowledge of community resources in four ways. First, they were asked to list the parenting resources that they were aware of in the community, and a count was made of the number of resources listed. For the other three community resource questions, participants were asked about a list of 21 social services in the community. For each agency, participants were asked if they had heard of it, if they knew what kinds of services it provided, and if they had participated in any of its programs. The number of yes responses to each question was tallied. The four questions described above did not seem to measure knowledge of community resources equally. For example, it is possible to know about resources without participating in them. For that reason, the participation question was rejected as not measuring knowledge of resources. In order to determine the most accurate measure of such knowledge, correlations were used to measure the degree to which the remaining 26 three questions measured the same construct. First, the number of agencies heard of and number of services known were significantly correlated (r = .61 , p<.01 at Time 2), although this correlation was not as strong as expected given the format of the questions. It seems that participants readily stated that they had heard of an agency, but they were hesitant to say that they knew the services provided by that agency. In fact, the decrease from number heard of to number of services known was approximately 50% at both 8 weeks and at 12 months (8 week interview decreased from an average of 6.46 agencies heard of to 3.68 services known, and 12 month interview from an average of 6.55 heard of to 3.01 known). Given this discrepancy between services heard of and services known, the latter was determined to be a more accurate measure of knowledge of community resources. The remaining variables, the free recall list of community agencies and the number of community services known, were significantly correlated (r = .38, p < .01 at Time 2 ). However, the free recall list included personal resources, such as family members, physicians, and books. Although these resources are valuable, they are not general community resources. Therefore, this item was rejected as a measure of knowledge about community resources. The remaining question, the number of community services known, was determined to most accurately reflect the knowledge component. This variable could range from 1-21, although in reality, interviewee scores ranged from 0-14 at 8 weeks (mean = 3.68, standard deviation =3.17) and 0-18 at 12 months (mean = 3.01, standard deviation = 3.68). 27 m o_f Bream math m film Hypothesis 1b states that the longer a participant is involved in the program, the better her support system will be. In order to measure the amount of support that mothers had, two different questions were asked. The first question was "Who do you turn to when you have questions, problems, or concerns about your child?" A follow-up question asked, "Do you feel that you have enough help to address your parenting concerns?" Thus participants made a free recall list of personal supports and made an evaluative judgement as to the adequacy of those supports. Each of these variables is further defined below. m if m m mentioned. Participants were asked to list who they turn to with questions, problems, or concerns about their child. The number of people they mentioned was counted. This number ranged from 0-6 at 8 weeks (mean = 2.28, standard deviation = .96) and from 0-6 at 12 months weeks (mean = 2.25, standard deviation = 1.11). Adeguacy pf resources. Participants answered either "yes" or "no" to the question about whether or not they felt that they had enough help to address parenting concerns. However, only 7 of the 159 participants who responded at 8 weeks answered "no" to this question, and at the 12 month interview, only 5 out of the 172 respondents answered "no" to this question. Therefore, there was inadequate variability to test the hypothesis that program length affected reported adequacy of supports. 28 Effects _o_f Low Intensity pp Participant Benefit Hypothesis 2 states that the more calls a participant received, regardless of program length, the more she will feel that the program helped her become a better parent. This hypothesis tested whether increased contact with volunteers increased the likelihood that participants felt that the informational packet was helpful (because the calls supplemented the learning from the packets) and that the phone calls were helpful. N_um_b_e_1: _o_f w mg. The KCHS parent education program coordinator tracked the number of calls participants received from volunteers. The number of phone calls ranged from 0 to 8 (mean =2.08, standard deviation = 2.22). Mg o_f informational kaeti The question "Has any of the information from the materials provided by KCHS helped you to be a better parent?" was asked, and parents responded with either "yes" or "no". Of the 86 participants who responded, 55 said yes and 31 said no. Therefore, each category contained enough responses to perform a statistical analysis. Role o_f phone calls. Participants also responded either "yes" or "no" to the question, "Has the telephone contact with KCHS volunteers helped you to be a better parent?" Forty-eight of the 94 respondents answered yes to this question, while 47 said no. For this question, enough responses were contained to perform a statistical analysis. Efifgpps o_f Parenting Exmrience 9p Participant gem Hypothesis 3 explores the role that participant characteristics can play in determining the amount of benefit gained by participating in the parent education program. Specifically, primiparous parents are expected to gain more from the program 29 than multiparous parents. Hypothesis 3a will examine the differences between parent groups in informational benefit from the program, and 3b will explore the proportion of parents who felt that they did not benefit from the program. Parenting exmrience. Information about the number of children in the participant families was recorded on the referral screening form, and it was this information that was used to separate primiparous from multiparous parents. _B_epefi_ts 9_f me program: sppmpt gel information. As part of the telephone interview, participants were asked to evaluate the telephone calls they received with the question "What did you frnd most helpful about these phone calls?" The answers given were originally coded as 0 (nothing listed or negative comments), 1 (comments about the supportive nature of calls, such as comments about non-specific help, caring, and fiiendliness), 2 (comments about information provided, such as developmental information, community resources, and factual answers), or 3 (comments including both support and information factors). The database contains a total of 103 responses to this question, and each grouping contained enough responses to conduct statistical analyses. In order to evaluate Hypothesis 3a, that a greater proportion of primiparous than multiparous mothers found the information to be helpful, original scores were recoded as either 0 if information was not listed (combining scores of 0 or 1) or 1 if information was listed (combining scores of 2 or 3). When the original variable was recoded, 76 of the 103 responses did not contain information and 27 of the responses contained information as beneficial. 30 To evaluate Hypothesis 3b, that more multiparous than primiparous mothers would state that they did not benefit from the program, the original scores were recoded as either 0 if positive comments were made (combining scores of 1,2, or 3) or 1 if no comments or negative comments were made (originally scored as 0). Recoding this variable led to a distribution of 70 positive comments and 30 negative comments. Results Hypothesis 1 stated that the longer someone participated in the KCHS parent education program, the more knowledgeable she would be about community resources and the more people she would name as parenting supports. In order to test the hypothesis that greater program length would lead to greater knowledge of community resources, a series of t—tests was conducted. It was anticipated that comparison mothers would be less knowledgeable than 8 week program participants as measured at 8 weeks. At 12 months, comparison mothers were expected to be less knowledgeable than 12 month participant mothers. Finally 8 week participants measured at 8 weeks were expected to be less knowledgeable than 12 month participants measured at 12 months. T- tests revealed that at 8 weeks, 8 week participants knew fewer community resources than comparisons (1(155) = -4.53, p < 0.001). At 12 months, 12 month participants and comparisons did not differ (t (126) = 0.49, p > 0.05). Finally, a comparison of 8 week participants at 8 weeks and 12 month participants at 12 months revealed no differences (t (129) = 1.26, p > 0.05) Table 3 provides group sizes, mean scores, and standard deviations for these analyses. 31 Table 3 Group Size, Meat; _a_ng Standard Deviations Q Analyses Variable (measure) p M (SD) Range Hypothesis l-— t-tests Information (Number of Community Resources Known) Eight Week Participants 80 3.94 (2.49) 0-8 Eight Week Comparisons 77 5.82 (2.71) 0-11 Twelve Month Participants 51 3.31 (3.16) 0-7 Twelve Month Comparisons 77 3.65 (4.21) 0-9 Support (Number of Support People Mentioned) Eight Week Participants 76 1.89 (0.78) 0-4 Eight Week Comparisons 77 2.65 (0.96) 0-5 Twelve Month Participants 49 2.29 (1.14) 0-5 Twelve Month Comparisons 58 2.21 (1.00) 1-6 Hypothesis 2 -- t-tests Did Packet of Materials Lead to Better Parenting (Number of Calls Received) Yes 55 3.73 (1.99) 1-7 No 31 3.26 (1.88) 1-7 Did Phone Calls Lead to Better Parenting (Number of Calls Received) Yes 47 3.64 (1.98) 1-7 No 47 3.68 (2.00) 1-7 In order to test the hypothesis that greater program length would lead to greater number of support people mentioned, a series of t-tests was conducted. It was anticipated that comparison mothers would mention fewer supports than 8 week program participants as measured at 8 weeks. At 12 months, comparison mothers were expected to list fewer supports than 12 month participant mothers. Finally 8 week participants measured at 8 weeks were expected to mention fewer supports than 12 month participants measured at 12 months. T-tests revealed that at 8 weeks, 8 week participants listed fewer support 32 resources than comparisons (t (151) = -5.36, p < 0.001). At 12 months, 12 month participants and comparisons did not differ (t (124) = 0.51, p > 0.05). Finally, a comparison of 8 week participants at 8 weeks and 12 month participants at 12 months indicated that participants in the longer program did mention more support resources (1 (123) = -2.29, p < 0.05). Table 3 indicates group sizes, mean scores, and standard deviations for these analyses. In Hypothesis 2, two measures were thought to be related to the effects of program intensity, as measured by number of phone calls. First, the effect of intensity on the amount of benefit participants felt that they received from informational packets was examined with a t-test. The results indicate that those who felt that they benefitted from the packets did not receive more calls than those who did not feel that they benefitted from the packets (t (84) = 1.07, p > 0.05). A second t-test was performed to examine the relationship between the number of calls received from volunteers and the feeling that the calls were beneficial. These results indicate that people who felt that the telephone calls were beneficial did not receive more calls than those who did not feel that they benefitted from the calls (1 (92) = -0.10, p > 0.05). Table 3 indicates group sizes, mean number of calls, and standard deviations for these analyses. It was hypothesized that participants who received more calls would feel that they received more benefit from the informational packets because the volunteers would refer to packet materials and explain more about materials contained within packets. However, only 18% of participants stated that they had read something in the packet that they did 33 not know before. Additionally, 19% of participants said that materials from the packet helped them to better understand their babies’ development. Hypothesis 3 examined the role played by parenting experience in benefit from aspects of the program. In order to test the hypothesis that a greater proportion of primiparous compared to multiparous mothers would state that they benefited from informational aspects of the program, a chi-square analysis was conducted with 38 primiparous and 61 multiparous mothers. The results did not support this hypothesis (x2 = 0.04, p = 0.85). To test the hypothesis that a greater proportion of multiparous than primiparous parents would feel that they did not benefit from the program at all, a second chi-square analysis was performed. The results indicated that the two groups did not differ (x2 = 2.02, p = 0.16). Discussion This study was intended to determine the effects of program length, program intensity, and parenting characteristics on a child abuse prevention program for low risk mothers. Our results suggest that the longer program did enhance program effectiveness in terms of number of supports but not in terms of knowledge of community resources. Additionally, the increased number of calls did not affect the benefit gained from the phone calls or packets. The fact that there was no relationship between number of calls and benefit from packets may be a result either of the fact that material was already familiar to participants or that these low risk women were sufficiently well educated to benefit from the packets without volunteer input. However, it is also possible that volunteers did not refer to materials in the packet and thus phone calls were not 34 additionally informative about packet materials. It is important to note, however, that the number of phone calls from volunteers was also not related to the amount of benefit that participants felt that they received from the program. Finally, parenting experience did not affect the amount of benefit that parents felt they received fiom either the packets or from the overall program. In general, participants in this study did feel that they benefitted from the parent education program. Based on participant comments, the vehicle of this perceived benefit appears to have been the relationship between participants and the volunteers with whom they had contact. Comments from program participants included the following statements: “I think this is a positive program. It’s nice to have a follow-up to written literature. Educated ‘well-off’ parents usually fall through the cracks; after all, what is out there for those of us who don’t qualify for social services?” and "I thought the program was wonderful, supportive, and very professional." Several mothers stated that the volunteers were helpful when other resources, such as family and doctors, were not. The mixed objective measure and subjective measure results found here mirror the mixed efficacy results for parent education programs found by Taylor and Beauchamp (1988) and Weinman and colleagues (1992). Taylor and Beauchamp (1988) found that in addition to gaining competency in current problem solving, parents who have been provided with educational and support services showed an increased ability to confront hypothetical future difficulties and a knowledge of the appropriate community resources to contact for assistance. However, Weinman and colleagues (1992) found no significant changes on a measure of realistic expectations of the future. These researchers suggest 35 that parent education programs produce gains primarily in concrete and specific tasks (Weinman et al., 1992). Based on this definition and our lack of difference in community resource knowledge (a concrete measure) based on continued program involvement, the current study might more accurately be labeled as supportive rather educational. The decision of what type of gain measure to use is driven by the goal of the evaluation and the collaboration between program personnel and evaluation personnel. In this case, the parent education program was a less intensive program than the home visitor program, with a shorter program length and fewer contacts between volunteers and participants. Therefore, it was decided that a less intensive evaluation should be conducted. Rather than using direct abuse outcome measures or established indirect objective measures of gain, this evaluation of the KCHS program utilized self—report questions. These questions are subject to the biases inherent in all self-report research measures, such as response biases and a tendency to avoid extreme scores (Copeland & White, 1991). Although participants were asked about their knowledge of specific community resources and were asked to list support resources, these measures may not have corresponded to child abuse prevention. Another more direct measure of child abuse prevention, such as referral to child protective services, might have shown a significant difference between participant and comparison families. There are several possible explanations for the lack of success of the KCHS parent education program. First, it is possible that child abuse prevention programs in general are not effective with low risk mothers (Pillow et al., 1991; Siegel et al., 1980). It may be that these mothers have adequate resources to address their needs without outside 36 intervention. A second possibility is that prevention programs can be effective but the KCHS program did not meet parents’ needs. It may be that another format (such as group support program, structured class situation, or home visiting program) is able to reach low risk mothers better than a less intensive telephone call program. Finally, it is possible that the KCHS program did create a difference in the lives of participant parents, but this evaluation was not able to detect this difference. Because of the fact that there is more than one possible interpretation of the failure to find results for this program, the picture is still unclear. Future research should take into account more comprehensive measures to clarify this discrepancy between objective and subjective measures of gain. One way to more effectively measure change due to a program would be to add pretest instruments. This would allow evaluators to measure the change within individuals from before the intervention to after the intervention. Not only would this help researchers determine who benefits most from the program and why, but the repeated measures design would also have more power to detect changes than a one-time comparison between groups (Oliver & Berger, 1980). However, Oliver and Berger (1980) caution that the addition of a pretest may increase the ambiguity of results because it introduces main effects and interactions, which complicates analysis of data. 1 Additionally, Arvey, Maxwell, and Salas (1992) warn that adding pretest measures will affect the relative power of the design when total cost resources are limited, as was the case in this evaluation. Concerned communities are very invested in child abuse prevention. Prevention program planners and implementers all believe that they are doing their best to strengthen 37 families and communities. This belief makes it difficult for people to question the success of prevention efforts. However, effective and objective evaluation is critical to determining whether or not child abuse can be prevented and which efforts work or do not work. 38 APPENDICES 39 APPENDIX A 40 3303.82: me 833me 22385 05 59> 03588 8 ~5me 8: 83698 05 use sec—became E038 38.8%: 05 can: 26: _ 38on 06 we coves—mg m2“ 5 23693 2 seems Egg—9, so.» 35 838 : .Eoumoac wage—Em 05 8398 S wcmooume .nm .Efizg b30388 mm 32385 £5 E 833693 50> .moocoscomaoo b3 “:05; 250% :3 so» .wctoBmSw 033.8885 _oom so» 8526 5:093 Em .04 805 .3 .88on gm magnum .0950 “com 05 Set 832% 233 0338 3 bang—o So» so menace o: 26: E? 8: .8 28:83 80.: 53mg 9 082.3 so» 5523 .38er8 Box on E? 32625 035 8 $0328 50> .3382 .3502 023 so.» .230 28%. 9732—8 05 v5 defiance? mo “8—er 05 8 @838 80:33 688 so» x3 3 85 2:03 03 doucotoufi vim—-32 Maggi tfim 3:31 .0550 Sov— ofi cues—«>0 3 So: 8 398:: 0.8 $833.5 wage—B.“ of. 9.9.0 net—SE8:— - 8.5...— 2.3an 3030.5 95.5 «835285 _ 08E. ”< 595%? 41 0000000 0000 00 0003000 000.» 00 030000 00 000000 00 00 00003 03.00000 .00 000—000 05 0030000 00.» 00c0 80.035 005 0.2003 05 .0000 0:00 000:0 0:30:00 030 05 00000 Ewsofi :0» 0003 305— 00 0003 00—0 0 3 .0000 0>00 E0000 0.02.. 00.» 006 000205 000 003 005 320.30 003 0005 a 000 .00» 00 000—00 0003 03000000 0005 300 305. 00 0003 0 .5 .0000on 005 .3000: 50000 000M 05 .3 00.» 00 0000 03000000 .00 000—000 05 00000 00000000 0800 00.» 0.00 00 w0m0w 300 E0 0 ”000000.005 “03:00 0.0303035 ”3030005 .00 0000 000802 000.30% "00002 42 _chidé-d $308 05 5 BEBE 8: 33 86 “cacao—gov 333 Box Scan 265— 8 8:— :o» 2:03 cougar—£5 go 396 :33 .v m: 33 8:3 .8» .3 A28 225V oz 3% 20:3 Havana—gov 333 So» _ESEoES so» Eon BEE x55 so» 85 .8 .29» 8 26: 83 “m5 H833 05 E nouns—:85 ha 955 33 .m a: 83 3:3 .8» .2 A28 29:8 oz 8.» $8.33 Sons 305. 8: Eu so» 36 so» 8 3:2: $3.538 mo “av—ca 05 5 9.2.35 Son—m v8.— 30» 2Q .N map—m tfim 3:on 3550 Box 05 3 so» 3 «new mirage .«o «8—08 05 Bonn Ema—o: ~88 am so» Ev was? ._ £5538:— ana-.soa— no nouns—gm “aha—5038.5 Adm 8—8 8: ow .5: 05 8 85mm :8: 05 080 288 can xocowa 883:8 85 8 03388 08 803.88 885 .8 as?“ 8: mm 2?. .02 mm 85 0.88 E83 8% 55 $88880 8388 00¢ 98 805 03288 $828 05 85:3 27. can 680802 .8 use: was 27. mg 85 mama 28088 a 68886 8m .8: 8 88.80 98 85 850:? 8 woman 52 .8 mm>v 8888888 8 888 88 8265 05 858 88$ .53 85mm 88388 a 8 28:88 moorcom 05 808 882% 888982 05 .3 am: 05 8 hocowa 80: 05 8 8 988 02 38 2:05 a .0m 8 8 cm mm 8 mm.» 8365 062?. .2 am: 05 8 mocowm 68: 08 8 8 288 02 $8 050% a .m» 9 8 cm $8003 8 £0: 8.“ 00 EB 00 08 0: 00: :0» 0023 8.“ 0500000 $500000 >00 00: 00» 0203 A08 2008 02 00> 0050080 3000.80 80% 0000000 00 E0: @0000 032 :0» 005 _00.“ 00» on “00:00:08 03000 .8 000802 300000080 00 m\&:m:000_0m .m— .N_ @300 30: 80» 5000 050080 8 £80308 £80800 05: 00» 00:3 00 EB 00» 00 0:? 080380880 0030000 8: was 000 00300000 8: mm 220 .U 080008088 0030000 was 220 .m 00300000. mm EEO .< A000 20:8 000» 08030888 00503 0020000 0.30 .80» was 8 80888000 00038088 “me 00:30 8.“ 220 80% 00300000 00.20 00% 083 A28 2008 02 00> 00000000 .8.“ 000030885 00¢ 000308 808.8003 530m €800 000M 05 :20 00030 :0» 0E .: .3 46 ”3:08:80 028500< 47 APPENDIX B 48 003008080 8 0.8885 0503.880 08 003 008800 00 00080 000 80000080 08 08 8080080 80800 008880 05 000.— 080 0 888000 05 .8 008888 00.0 8 080000.80 00 0088 508088., 00% 88 8008 0 .8888 9:30:00 05 00388 00 8000.80 .5 $88200, 0008—0800 00 300.8080 000 8 808000080 80> 00000000800 88 80085 000000 80 00.» .8838 0008808008 _000 :0.» 8000000 8800.80 08 00 0.85 .2 8800800 0000 00 .03 8000000 0005 00 08388 80> 080080 808080 80% 08$ 800000 8.0 EB 00.» 00003 08 808020.000 0.0000 80.» .00080000 @8888 .8 800—305— 80.» 00 008000 8000000 0800 :0» 000 00 00: 080B 0 3 8080808 0000.30— 0.88w000 88m .9080: @800 80% 05 08:88 8 0000 00 0000080 08 8000000 8030—80 000. 0809 80800800 - 8.8..— 80800 800800 00000 8080800 0 0800. um 80000044. 49 0000000 0000 00 0.003000 80» 8 0000000 00 0000000 00 00 0000—0 0800000 8000000 800 0005 0.00 00» 00003 8000 000 5088800 80» 8 00080000 8000000 8 000000030 80» 8000 00000000 0800 00» x00 00 w80w 300 80 0 ”000000.008 ”8000: 0.8388080 ”3003080 8 000D ”000082 88.8w ”0802 50 Adm x3 8: cc .3: 05 :o macaw“ as: 05 8:0 o>oE Ea xoaowa 3:5qu 35 “a via—Raw 2a moomtom :33 mo 29% 8: fl 2?. .OZ 8 35 088 2:95 no» 55 $53280 5353: 02m 2a 205 03282“ 8338 05 35:3 2.7. 93 £33on mo Emu: mm: 27. mm; 35 "£8 28088 .2 63883 8m .8: .8 8280 0.8 55 850:3 no woman 6 Z 8 mm>v 5:33:23 5 838 28 33.0.5 05 :38 8:3 .83 hog—own 83033 a am 03236 8228 05 Sons 332% 282898“ 05 .3 am: 05 :o macaw“ :8: 05 8 .8 968 0 Z 99% 0E2? .2 .0m 3 co ow mm 3 mm; £038“ 060% .2 .m: 05 :o magma :8: 05 3 co 968 CZ 993 2:2? : .mm 2 :0 cm &E .2 tfim 5>m .2 :0 ram .: 08:0 852:6 EEO mEamM @520 .2 5am 932 .o “_o_mcoaxm mOéU .0550 Euv— .w 0:5 3:0 2835500 .5 852 .283 23:; .o moomtom 080 5:8: moi—«62$ WEBEEU .m OE .v zocnoo 8508M bung Ea EEO .m @5533 3.222 a fix €13: .N 28602 A NHOZ _nma muOZ _nmm> NHOZ _ummcw was: uofifiomtmq ~58 03256 Em wimo Ede: .55 so» 983 803.3% 3:3 >55— soz on 56 so» 025 .Um .mm .5 was so» 96E .w A28 29:9 02 mo.» $50280 wuss—2mm So» $233 3 Be: 5325 033 so» 35 30.“ so» on .N. 628358 Bacon mo 89:32 ”Houseman: 8 mEEmcota—om ~33 33 So» Sosa 35280 S £8032; $5383 05: so» :23 9 EB no» ca 055 .0 20:35:85 3382 8: mm: 28 3162—8 8: fl BBQ .0 maozmficzafim 3382 8: BED .m 3:628 mm EEO .< A28 29:8 So» mcoumnisfiem 56mm: 3332 230 So» was .5 acogfioaqa 8:35.55 “Em 55mm: SM 230 Box 31638 $56 so» 035 .m A28 286v oz mo> $2220 Sm acoumnmgeg out 8235 608530 530m 5580 28M 05 35 983“ so» 8< .v 53 “58888 3822:. 54 APPENDIX C 55 550 a 89298.8— :288888— a cows—885 0.30 BBQ 6 A90 £28638 38088—33 rmdv 83880.8— _SEEQBgQQ Am 680 women—own no 8388 082322 rwdv mooBOmom ban—8800 .«o 53 A< $8.03 80% 8 a: x02 3938.8 so» ow 8038888 .«o 39¢ :25» .v 838088 8: on 88: mo ~38: Z $30M 3 so» 8 80m matey—W8 05 8 v2.32 so» 26: .89» ”man 05 mats—u 88: 858 303 .m C. 823.28 9 8% .N 28 _ 80528 58 8 02 80265 83833.. ,5 ~82? 88% “38088 no» ow .8» «2 A28 285V 02 mo> $30M 89¢ Ho: 83 85 So» “ma 2: mamas 838.838 @58an 8a 3382 so» 26$ .m A28 225V 02 £5 w88mo£ 88m 3:3: fiasco 80M 05 88a 8288888 mo “8.03 a 83608 “38088 no» ow .89» 63 05 8 08: 8w 2 .8: 96: 8050 EB 88on 83m 8283 3880 Box 05 88% 8088888 mo 38—03 3382 26: $5.89 083 ._ 8:89:85 amass?— ua 858.55— 9.; 23> "9:88:890 ”u “8.3% 88on 32235 N 08C. ”0 5833‘ 56 A8888 8 88 88 088 8 .888 a 88 88 m8— .888 a 88 888 8 83.. 828 »n 8888 88 .83 88 88 8 5 $8882 888088 8» 8 8—8 088 »88 28: .8» .8 .mm A2 8888 3 82m 82 8838 88:88.. .5 88 288 oz 8 > “88889 8m 888$ 5:80 838 888 8—8 088 »8 8388: 888082 8» 8 :8» 88: 08 885 .8: 26: 888 88 88883 8m 888m b880 808 8:8 88 088 8388 2:8 $88: 088 .w Addévfi m8» 3 8888 :8: 8: 8s 88 8088888 $38 :8» :88 >85. 8 8:— 8» 28>» 88888 .8 8:»8 8.5» .N. 88:8 .8» .2 88 0.88 oz 8% 9:28 888 a B 8 8» 888 mEUM »9 888.8 88.888 08 888 88888 05 .8 »8 8m 8 A“: 83 8:3 .8» .8— 88 280V :2 8? N. 838 8088—088 @280 :8» 88888 8 8» 8:8: 85 8 288 :88 >85— 8: 88 8» 88 mmUM 8:8 8» 9 80:88 88888 .8 888 08 8 w88»:m :88 w8w8: 898888 8» on .m 57 wan—08 0.88 8009 08: 258 0888 05 8me 3cm .3 $.58 088 085 585» 5580: 588 5:5 8» 85 853 .o $580,508 085888 088 83 .88588 88» 80388 3580553 8 8:8 88 082305— 8 88. 88 .8 £805 8 68% 8» 883 55> 800889» mmUM 05 8 885888 88» 083 853 .3 .838 88» 8898 0803 A08 08.88 oz 8 > 88.8 »05 8 5080.88 0.83 8088—085 £580 88» 588 80588 88» 8 =8 85 om 30858.88 5085 858 05 0803 .0» 85 85 8» »83 889$ 080$ I585 83. I085 88:0 8 2 I085 80:8 8... 8:8 800389 5088 85 085 8 888 05 8 x85 8» 85 855 Am 58 Adm— x8 8: c: an: 05 :o 5:0»: :8: 05 3:: 96:: :5 3:09“ 81:53: 35 a: 03am?“ 8: 8238 «an? a: 2:3: 8: mm 27. .OZ 8 85 0.88 2:95 :9» :05 $53330 555:: 02% 03 205 29256 832% 2: 3.623 0% “E: £3232 mo :53: man 0% mm; 85 mmam 0:888 a 638.88 :om .8: 8 69:8 2: >2: 550:? :o woman 52 :o mm>v 53:58:80: : 83:: :5 826:: 05 :30: 8c? Amm G 3:09“ 830:3: : a: Ban—mg: $233. 05 Son: 832% 83:38.. 05 .3 am: 05 :o 5:03 :8: 05 3 :o 05:: 0 Z 9:3 052% a .02 8 :0 cm m2 8 mm; $038: 060% .3 am: 05 :o magma :8: 2: 3 :o 32: CZ 993 050% .2 £2 8 :0 cm <2 8 mm.» @532: 20233: .2 ”83328:c 5.882. 3:03 .8333: some mo 38: :96 96: so» .5523 2: =8 8 :9» 8—: 2:03 H 93 55888 :5» E 86:03 83:8 308 “5.8.5: Son: 2:» gm: 3 wfiow :8 :62 .2 $3332: 69:52 mbmgéoo .59» :_ .«o 2:3: so» 2: 82:88 3:53: 355 .S :398 do» .2 3:: 0:58 02 mo> 9:96: 3:3 : on 3 so» won—u: E3533 $5M .33 68:8 33:28 05 $5 .2 59 059004 050040; ._N game: 5% .8 _Sfimom mama—2 am Hm. 0.0085 @553.» tfim EEm .2 as: 220 332280 £56580 0». 980 so: .2 0.8005 588% Ea ESE - 3:83 mfitoamzm 3:03; .2 Hmm .2 Ammmoolxv 005:0; coon .2 moomZOm NASSZEEOU Soy—.50 x"; Elna: .2 5% :0>m .m— =o ram .: 256 852:0 220 fig: 280 .2 imam @003 .o :omm:00xm @060 .0550 Eovm . 0:5 2th 0:895:80 . 352 £8: 255 . US, . €550 8:58: SEE Ea 220 . w n c m0oE0m 030 530: “SE—060% @5620 .m w M 305535 :awfouz :_ mag b.2003 .N _ 28602 . mnOZ _nmm; :55 “0033250: :0>0 so» 020m .02 NHOZ Tmm> as“: 0302?: 03 00820.0. “0:3 393 so» 0Q .92 NHOZ ~nmm> page 200: :0>0 so» 0.63 3:2 60 ”Ea—:08 .00.» .: »0:o 08:9 02 00 > 88:30 :0 :8: .8: 8 53 0: 0:0 0: :0: so.» :03? .8: 0:80:00 33:08: .»:0 :0: :o.» 0.6: 9:0 0.05% oz 00 > 00:80:00 33:08: So.» 000028 3 :8: $35 0.6: :o.» 8:: _00: :0.» on 60:03:08 2:00: :0 03:52 5:030:00: 9 0\:E0:o_8_0,: 0230 :8» Son: 0:80:00 :0 0:838: £53003 0%: :9» 8:3 9 53 :0.» o: 055 3:0 0:005: 02 00 > 005300588: .8503 :o 08: 8 :: .»=:0h:0 0: 230 :5.» 85 0.6209 :0.» on 33800: a :85 0:56 0000: 8 “00.0800: 230 So.» 00: 08:00:58.5 .»::E 33: .m— .2 .2 .m— .3 61 ”00:08:80 8:032; 62 APPENDIX D 63 8303:08— .:0 0808mm 05030.88: 0:: 335 0::3:00 0: 30080 00: 80:83.8: 0:3 3:0 80800.80 8088 3088.3: 23 300: 0>0: _ .8088: 0:: :0 830208 0:3 8 0308080: 0: 00.80 b30830.» :0.» :05 08008 : .0:0300:: 835030.: 05 830:0 0: 88080 .»m ..»:0:8:0> .»_0:0_:800 0: 30388: 033 8 :030:83:0: ::0> 0008100800 .»:0 80:33 0::003 :00 :0.» 3:30.580 0308038008: :00: :0.» 8300:: 8303.8: .»:0 0: 08:: .3 .8088: 803m .»::003 .»:::00 80M 05 88.: 0003.80 0:88 03000: 0: 33338—0 80.» :0 8300: 0: 0>0: :3: :0: :0 80300:: 000:: 830:0 0: 88:0 :0.» 850:? 83:33:00 ::0: 0: :3: 80300:: 000:: 0: 08380 ::0> :00.» :00: 0:: :0>0 303000: :0.» 0:00 0:0:: 8.3030: 0:: 3:0 83080.8: :0 8:00: 0:: 0: 30:08: 80300:: 0800 :0.» :00 0: 0:: 2:03 0 3 8038808: 0.03.30: 0.80.88: 85m .»::003 .»:::00 80M 0:: 80308 0: :8: 0: 3038:: 0:0 80300:: 835030.: 0::. ::0:mv 83:90.83:— .. 8:0: 80085 80:m :0>0U 3:0 8:0 ”— 3:00:00 303.88: 838808: m 08:. ”Q 0:38:90: 64 80300:: 3000 0: 80380 :00.» 8 033000: 00 03:00:0 00 03 0000:: 0:008:30.» 80:80:: 80:8 35:00:: 80:: 0300 0:03: 05 8030 88:05 :0.» :033 30:0: 0: 803 3:800:00 0 3 8003 038 38030 3:33 :0.» :05 8080:: 05 8 303808 :0: 003 :05 885.80 003 0:05 :3 3:0 .:00.» :00: 05 :0>0 :0.» 0: 8:803 :003 003 8080:: 0:5 303 305: 0: 803 0 .3 8080:: 30:8 35:00:: 5:000 83: 05 8030 00030008 0800 00.» 0:00 0: 8808 30: 80 : ”0:030:38: ”30:0 ..»030 88.68 00 3000 83330 :080: :0 :0::0 :00.»0:: 0: 0:3 :00: 808033: .80: :305: :: 080:0 8088008 :0: 800003 A 0300 :0 :03805 :03:0>:08: 82:88:: 03808 N: 03003 8 8880208: :0 58:03 ”0:038: 0.838808: ”3320:: MD 0:89 000 5:0 0.220 03:52 258 U2:02 0.220 02:52 08.300 000: 082: 0:02 65 APPENDIX E 66 :030:>:08: :0 0:80:85 303:0:8 05 5:3 0:8800 0: 30080 003 80803.8: 05 3:0 8080080 ::00:00 30888 05 300: 0>03 : 8080:: 05 :0 830803 033 8 0:080:80: 0: 0080 .»3:0880> :0.» :05 08008 3 .0:0300:_0 883030: 05 :030:0 0: 880080 .»m ..»:08:_0> .»_0:0_:800 0: 308:0:8 033 8 8308038: 80> 00000800000 .»:0 805:3 0:3003 :00 :0.» 888380 0380:8008: :00: :0.» 00300:: 833:0: .»:0 0: 0:05 :: 38803800 803 03 :33 80300:: 0005 0: 80380 80> 080800 88820: 80.» 5:3 80:80 :0: 8:: :0.» 0:033 3:0 .808:0_0>03 0.3330 80.» .0008000: 3388800 :0 0830—305— 80.» 0: 30:80: 00030080 0800 :0.» 300 0: 00:3 3803 0 3 .:03:0>:0:8 0:03-30: 0.8088: 888 35:00:: .»:::00 80v: 05 0:0:_0>0 0: :303 0: 303:0:8 0:0 8030080 883030: 03:. ::0:0 0003:8800 .. 8.8.: 80800 :003m :0>0U 3:0 8:0 : 800:00 303:0:8 8030:800 N 08 H ”m 033:0::< 67 .:8001: 3000 0: 80380 :00» 8 033000: 00 0:300:30 00 03 0000—: 0:50:00 8880.5: .508 533 :8: :0» 0:033 8030 3:0 8808800 :00» 8 0008000: 88820: :0 000:0:030 :00» 8030 0:500:30 0800 :0» 300 0: 8808 30: 80 3 “0:050:58 ”0308:: 080303208: ”303208: :0 0:0Q 000 5:5 0.080 003802 0:03: ”0802 0.3330 0352 60.300 €000 0:80 ”02.02 68 LIST OF REFERENCES 69 LIST OF REFERENCES Abidin, R. R. (1990). Parenting Stress Index Manual. Charlottesville, VA: Pediatric Psychology Press. Arvey, R. D., Maxwell, S. E., & Salas, E. (1992). The relative power of training evaluation designs under different cost configurations. Journal 91‘ Applied Psychology, fl, 155-160. Barth, R. P. & Ash, J. R. (1986). Identifying, screening, and engaging high risk clients in private non-profit child abuse prevention programs. Child Abuse g4 Neglect, 19(1), 99-109. Briere, J. (1988), Multivariate correlates of childhood psychological and physical maltreatment among university women. Child Abuse 11151 Neglect, _l_2 (3), 331-341. Caldwell, R. A. (1991). Mediational screening: Is the benefit worth the cost? American Journal 91' Community Psychology, 12 (6), 847-851. Cohn, A. H. (1979). Effective treatment of child abuse and neglect. Social Work, 24 (6), 513-519. Cohn, A. H. & Daro, D. (1987). Is treatment too late: What ten years of evaluation research tells us. Child Abuse an_d_ Neglect, l_l_ (3), 433-442. Copeland, A. P. & White, K. M. (1991). Studying Families. Applied Social Research Methods Series, 2_7, Newbury Park, CA: Sage Publications. Goldberg, W. A. & Michaels, G. Y. (1988). The transition to parenthood: Synthesis and future directions. In W. A. Goldberg & G. Y. Michaels (Eds), Ih_e mifion Lo pmnthood (pp. 342-360). New York: Cambridge University Press. Gottlieb, B. H. & Pancer, S. M. (1988). Social networks and the transition to parenthood. In W. A. Goldberg & G. Y. Michaels (Eds), Lh_e transition _tp parenthood (pp. 342-360). New York: Cambridge University Press. Guterman, N. B. (1997). Early prevention of physical child abuse and neglect: Existing evidence and future directions. Child Maltreatment _2_ (1), 12-34. Harper, R. & Balch, P. (1975). Some economic arguments in favor of primary prevention. Professional Psychology,_§ (1), 17-25. 70 Krugman, R. D. (1995). Future directions in preventing child abuse. Child Abuse an_d Neglect, 12 (3), 273-279. MacMillan, H. L., MacMillan, J. H., Offord, D. R., Griffith, L., & MacMillan, A. (1994). Primary prevention of physical child abuse and neglect: A critical review, part 1. Journal gt: Child Psychology _8_L, Psychiag, 3_5 (5), 835-856. Martin, M., Scott, J ., Pierron, J ., & Bauerle, B. (1984). The Kansas family and children’s trust fund: Funding prevention programs in the eighties. Child Abuse app! Neglect, 8 (3), 303-309. Milner, J .S. (1980). 11;; Child Abuse Potential Inventory: Manual. Webster, NC: Psytec. Olds, D. L. (1988). The Prenatal / early infancy program. In R. H. Price, E. L. Cowen, R. P. Lorion, & J. McKay (Eds), Fourteen Ounces pf Prevention (pp. 9-23). Washington, D. C.: American Psychological Association. Oliver, R L. & Berger, P. K. (1980). Advisability of pretest designs in psychological research. Perceptual an_d Motor Skills 51 463-471. 9—’ Pillow, D. R., Sandler, I. N., Braver, S. L., Wolchik, S. A, & Gerstan, J. C. (1991). Theory-based screening for prevention: Focusing on mediating processes in children of divorce. American Journal pf Communig Pachology, _l_9 (6), 809-836. Reppucci, N. D., Britner, P. A. & Woolard, J. L. (1997). Preventing child abuse and neglect through parent education. Baltimore: Paul H. Brookes. Resnick, G. (1985). Enhancing parental competencies for high risk mothers: An evaluation of prevention effects. Child Abuse and Neglect, _9_ (4), 479-489. Rosenberg, M. S. & Reppucci, N. D. (1985). Primary prevention of child abuse. Journal o_f Consulting apd Clinical Psychology, $6), 576-585. Salzinger, S., Kaplan, S., & Artemyeff, C. (1983). Mothers' personal social networks and child maltreatment. Journal _o_f Abnormal Psychology, _9_2, 68-76. Siegel, E., Bauman, KCHS. E., Schaefer, E. S., Saunders, M. M., & Ingram, D. D. (1980). Hospital and home support during infancy: Impact on maternal attachment, child abuse and neglect, and health care utilization. Pediatrics 66 (2), 183-190. Taylor, D. K. & Beauchamp, C. (1988). Hospital based primary prevention strategy in child abuse: A multilevel needs assessment. Child Abuse QC! Neglect, 12 (3), 343-354. 71 Wallach, V. & Lister, L. (1995). Stages in the delivery of home-based services to parents at risk of child abuse: A Healthy Start experience. Scholarly Inguig fpr Nursing Practice, 2(2), 159-173. Weinman, M. L., Schreiber, N. B., and Robinson, M.(1992). Adolescent mothers: Were there any gains in a parent education program? Family a_n_c_l Community Health, _15 (3), 1-10. Westrnan, J. C. (1994). Licensing Parents: Can we prevent Child abuse and neglect? New York: Insight Books. Whipple, E. E. & Wilson, S. R. (1996). Evaluation of a parent education and support program for families at risk of physical child abuse. Families i_n Socieg, 11 (4), 227-239. Zigler, E. & Black, K. B. (1989) America’s family support movement: Strengths and limitations. American Journal o_f Orthopsychiag, _5_4_ (1), 6-19. 72 nICHIonN STATE UNIV. LIBRRRIES llWill“WI”WWlllllllll”11111111111 31293017197348