Why special populations are not the target of family preservation services: a case for program reform.

Curtis, Carla M.

The number of children who have been placed outside their homes of origin as a result of abuse, neglect, delinquency, emotional problems, or developmental disabilities, is astronomical and steadily increasing. Of this number, "special populations" like children of color continue to be disproportionately represented. Intensive family preservation, a program that attempts to reduce out-of-home placement rates, has not demonstrated empirically, a sustained record of success in the reduction of placement rates among special populations. The purpose of the current study was to understand the manner in which special populations are targeted for services by examining the attitudes, beliefs, and behaviors of a national sample of family preservation workers. Results indicate a significant bias against targeting family preservation services to special populations in general, and children of color in particular. Specific recommendations about the targeting of special populations are given.

**********

The reality that there exists extraordinary numbers of children in out-of-home care within the U.S. child welfare system is now common knowledge. Nationally, at the end of 1999, there were 550,000 children in out-of-home care (Administration for Children and Families, 2002). The situation specific to "special populations" like children of color is even more bleak, given the fact that they are disproportionately represented across the service continuum. In fact, children at most risk of remaining in substitute care for extended periods of time are children of color (Black Administrators in Child Welfare, 2001; Gustavason & Segal, 1994; W.K. Kellogg Foundation, 1995). Additionally, other special populations like "older children," "sibling groups," and the "emotionally disturbed," have experienced a rise in their out-of-home placement rates. The Administration for Children and Families (ACF) reports that over 120,000 of the half-million children currently in out-of-home care cannot return home safely because of their special needs.

Although many discussions have taken place concerning the need to remedy the overrepresentation of special populations within the child welfare system, a significant, programmatic address was not hoped for until the implementation of intensive family preservation services. Intensive family preservation services are short-term, home-based, family-centered programs that provide therapeutic intervention as well as concrete services to families who are at risk of losing their children to out-of-home care. Formal family preservation services began to surface in 1974 with the introduction of the Tacoma, Washington Homebuilders Program (Kinney, Haapala, Booth, & Leavitt, 1991).

Currently, there is widespread disagreement concerning the efficacy of family preservation programs. In January of 2001, the U.S. Department of Health and Human Services released the findings of major evaluations of family preservation that were conducted in New Jersey, Tennessee, and Kentucky. All three programs reported little program effect in reducing out-of-home placement rates and improving overall family conditions. However, concurrent with the release of the government's evaluation of family preservation services, Kirk (2000) released his final report of a retrospective evaluation of North Carolina's family preservation services. Contrary to the government's evaluation, Kirk found that not only were services effective, but previous studies have not provided evidence that there is a lack of treatment effect in family preservation services. Kirk concluded that previous studies have been unable to detect treatment effect because of the lack of practice wisdom employed in the design of evaluations. Despite the debates, fiscal year 2000 funding for family preservation services was $295 million (ACF, 2002). For fiscal year 2002, the Promoting Safe and Stable Families amendment (sub-part of the 1997 Adoption and Safe Families Act which guides family preservation service funding) will be funded at $375 million (NASW, 2002). Although there is an increase in funding from previous years, there is a caveat. Funding levels are not guaranteed as they have previously been. Moreover, the funding continues to support four programs: family preservation services, community-based support services for prevention, time-limited reunification, and adoption services.

Family preservation services are critical to the child welfare system and they remain one of the largest prevention entities within the system. Given the fact that special populations such as children of color are most affected by the child welfare system, prevention programs such as family preservation should make these groups the focal point of service delivery. However, research has clearly demonstrated that children of color are not likely to receive prevention services (National Association of Black Social Workers, 1992; NBCDI, 1989; Pinderhughes, 1991; Stehno, 1990).

This article includes the findings of a study that sought to uncover reasons why special populations do not receive prevention services, such as family preservation, at the expected rate. In doing so, the aim of this study was to seek the impressions of a national sample of family preservation workers concerning their experiences in targeting services to special populations. Although the concept special populations can and does include a range of groups, this article focuses specifically on children of color due to the differential service bias against them which was a key finding in this study as well as many others (Courtney, 1994; Neuspiel & Zingman, 1993; Roberts, 2002; Stehno, 1990; U.S. Dept. of Health and Human Services, 1997).

Family Preservation Under Examination

Service Efficacy

As of late, family preservation services have come under close scrutiny because of mixed reports concerning program success and also because of the public attention given to several high-profile "system" tragedies. Researchers, both internal and external to the philosophy of family preservation, have debated the program's merit (Berliner, 1993; Maluccio, Pine, & Warsh, 1994; Nelson, 1994; Pecora, 1994; Rznepnicki, 1994; Seader, 1994).

There appears to be some agreement that if family preservation programs are not having their intended impact, it is because of six main reasons: (1) inappropriate children and families are the target of services; (2) treatment models and modes of intervention are incompatible to client needs; (3) workers do not espouse nor demonstrate program values; (4) legislative guidelines are not being followed; (5) measures for determining client appropriateness are flawed; and (6) ambiguity is increasing concerning what constitutes an effective outcome and how the outcome should be measured.

Why Target Special Populations?

As family preservation decision-makers begin to rethink future program direction, the plight of special populations warrants attention. A significant factor in the decision-making process employed by family preservation programs should be the practice of targeting special populations. It can be argued that special population families experience a disproportionate number of hardships which can lead to their children being placed outside the home. Such hardships place special populations in a precarious situation, requiring the services of family preservation programs. In the child welfare system (ACF, 2002) as well as mental health, the following groups have been identified as special populations: children of color (Gustavasson & Segal, 1994), homeless children (Douglass, 1996), sibling groups (Drapeau, Simard, Beaudry, & Charbonneau, 2000; Smith, 1996), children of the mentally ill (Coiro, 1998; Finzi & Stange, 1997; Luntz, 1995), reunification families (Frame, Berrick, & Brodowski, 2000; Fraser, Pecora, & Haapala, 1991; Smith, 2000; Sudia, 1982; Talbot, 2001; Walton, Fraser, Lewis, Pecora, & Walton, 1993), potential adoption disruption families (Triseliotis, 2002), older children (Sedlak, 1997), chronic juvenile offenders (Loeber & Farrington, 2000; Scherer, Brondino, Henggeler, Melton, Gary, et al., 1994), severely emotionally disturbed--SED (Jonson-Reid, Williams, & Webster, 2001; Solnit, Adnopoz, Saxe, Gardner, Fallon, 1997), children under the age of five (Berrick, Needell, Barth, & Jonson-Reid, 1998), first-time parents, perinatal parents, children with birth abnormalities and/or the medically vulnerable (Berthier, Oriot, Bonneau, Chevrel, et al., 1993), children who are HIV positive or those who have AIDS (Anderson, 1998; Goicoechea-Balbona, 1998; Tenner, Feudo, & Woods, 1998), and status offenders (Nugent, Carpenter & Parks, 1993). More recently, children whose parents are infected with HIV/AIDS have been identified as a special population (Cameron, 2000; Draimin, Gamble, Shire, & Hudis, 1998; Mason, 1998; Taylor-Brown, Teeter, Blackburn, Oinen, & Wedderburn, 1998). The literature is most developed around the special population group, children of color. Children of color, enter the system at greater rates, remain longer, and experience more difficulty with permanency planning. However, what is neither fully known nor understood is the role of the special population criterion in selecting families for family preservation services.

Decision-Makers, Gate-Keepers and Federal Legislation

Arguably, key decision-makers and gatekeepers involved in the process of establishing criteria for the selection of children and families for services include: (1) legislators; (2) judges; (3) program managers; (4) referral agents; and (5) workers. Each of these five gatekeepers has a distinctive role in helping programs to realize their legislative goals, one of which is the targeting of services to special populations and other high risk groups.

Understanding the attitudes, beliefs, and behaviors of family preservation workers towards special populations may be facilitated by a review of the policy or legislative foundation that informs their practice. Child welfare practice is grounded in policy that is established through federal and state laws. Change in relevant statutes and subsequent change in practice is often influenced by demographics or demand for services, funding restrictions, or changes in appropriations, as well as philosophical shifts. One such philosophical shift can arguably be associated with the passage of the Adoption & Safe Families Act (ASFA) of 1997, P.L. 105-89. For children in the out-of-home care system there has been a shift away from family preservation aimed at supporting, if not reuniting, a child(ren) with the custodial parent(s), to a "child-centered goal" of permanence to include long term foster care, guardianship, or adoption as appropriate.

Family preservation has been discussed as both a philosophy which guides practice and as an approach to permanency planning for children in the out-of-home care system (Downs, Moore, McFadden, & Costin, 2000). The statutory foundation which supports and/or limits family preservation as both a philosophy and approach to permanency planning follows.

Special populations under P.L. 96-272. Some practitioners attribute the introduction of family preservation as a philosophy in service planning and development to the passage of the Adoption Assistance and Child Welfare Act of 1980, P.L. 96-272, and to the related research and demonstration projects that informed the passage of the legislation (Allen, Golubock, & Olson, 1983). Under this legislative mandate fiscal incentives at the federal level were directed toward substantive reforms aimed at ensuring permanent families for children. While the phrase "family preservation" is not in the statute, P.L. 96-272 was crafted to address the statutory goal of providing preventive and reunification service programs aimed at ensuring permanent families for children through the Child Welfare Services Program, Title IV-B of the Social Security Act (U.S. Code Congressional, 1980). Legislative history affirms that Congress intended for states to provide a range of services to prevent foster care, or ensure that reasonable efforts are made to return the child to the family, if appropriate, thus preserving the family unit (U.S. Code Congressional, 1980).

Prior to the passage of this legislation and the introduction of the reform measures aimed at family reunification and preservation, federal policy often encouraged the inappropriate separation of children from their families. States receive a substantial federal payment through the Foster Care Program, Title IV-E of the Social Security Act, to augment the costs of caring for children who are removed from their homes when allegations of child maltreatment are made, thus the incentive. The introduction of the "battered child syndrome" in the 1960s and heightened awareness of physical maltreatment, resulted in increased numbers of substantiated victims of child maltreatment and of those children who were removed from their home on a "temporary" basis. These factors contributed to the large number of children in the out-of-home care system during the 1970s (Wang & Daro, 1997).

During the 1970s a number of research initiatives identified a frightening trend of growing numbers of children being removed from their homes on a temporary basis then remaining in out-of-home care for long periods of time. This phenomenon referred to as foster care drift, overwhelmingly and disproportionately effected the special population children of color, specifically African Americans, Hispanics, and Native Americans. Other effected groups include sibling groups and children with physical and emotional challenges. The aforementioned research would provide the foundation for reform measures contained in The Adoption Assistance and Child Welfare Act of 1980, P.L. 96-272. The victims of foster care drift and those for whom systemic barriers prevented permanency were identified as "special needs children" (or as the strengths-based literature and more contemporary practice jargon says, "special populations"). The legislation called for unique efforts to promote permanency for this population (Allen, et.al., 1983; U.S. Code Congressional, 1980). One of the more controversial requirements was that in every case:

 reasonable efforts will be made (A) prior to the placement of a
 child in foster care to prevent or eliminate the need for
 removal of the child from his home, and (B) to make it possible
 for the child to return to his home (Department of Health and
 Human Services, 1996).

Critics of the reasonable efforts standard maintain that services which tend to help keep children with their families, when the family is unable to provide adequate care, place the child(ren) at unnecessary risk.

Statistics would suggest that the goals outlined in P.L. 96-272 were realized in part as the number of children estimated to be in the out-of-home care system in 1972 was 500,000, but by 1982 that figure was down to approximately 262,000 (Spar, 1993; Wang & Daro, 1997). This trend did not continue and by 1995 the caseload spiraled back to approximately 494,000 (Spar, 1993; Wang & Daro, 1997).

There are a number of reasons why the service goals of returning children to their families or keeping them from foster care were not sustained. States' initial resistance to broad sweeping system change and scarcity of resources resulted in minimal state program reforms. Alternative service or program initiatives introduced and replicated after the passage of P.L. 96-272 suffered from major cuts in federal funding in 1981, combined with heightened demand for service shortly after passage of P.L. 96-272 (Allen et al, 1983; Spar, 1993).

Special populations under P.L. 103-66. The continued increase in the number of children referred for child protective service precipitated the introduction of a new federal program in 1993. A program of family preservation and family support services was proposed as part of the Omnibus Budget Reconciliation Act of 1993, P.L. 103-66. The child welfare system nationwide was once again reeling from increased demands for protective services (e.g., the crack cocaine epidemic is partially blamed for the precipitating crisis in child welfare, other factors include poverty, homelessness, AIDS, mental health issues, other forms of alcohol and substance abuse) while also confronting high staff turnover and low morale, and a shortage of related support services such as drug and alcohol treatment and mental health care (Spar, 1993).

This program for preventive and supportive child welfare services for families is authorized by Title IV-B, Subpart 2 of the Social Security Act, and requires that states eligible for a share of a "capped" or limited federal appropriation must devote such funds to family preservation and family support services (U.S. Code Congressional, 1993). According to federal statute, family preservation services are intended for children and families in crisis or for families at risk of losing their children to foster care while family support services are intended to prevent crises, such as child maltreatment, from occurring.

Despite this legislative mandate, federal dollars have been used historically to support children after they are placed in foster care as opposed to providing services for families to prevent placement in foster care (Allen, et al, 1983; U.S. Code Congressional, 1993). For some states, placing emphasis on prevention and support services with federal funds represented a major departure from traditional practice (Spar, 1993). In fact, many states which began developing innovative approaches to helping families avoid losing their children to foster care, including family preservation services, relied on nonfederal and private resources (Spar, 1993).

When P.L. 96-272 was written, "special needs" children (more conventionally referred to as "special populations") included those who were at increased risk of being removed from their home due to allegations of child maltreatment. These children were and remain today disproportionately poor, Black, Hispanic, and Native American (Stevenson, Cheung, & Leung, 1992). Likewise, because federal family preservation and support service funds were also earmarked for special populations, they should be targeted for these services.

Without engaging in the "muddy" and emotionally charged discourse concerning the effectiveness of family preservation and support services, including the argument by many that the prima facie issue is poverty and the impact on its victims, the focus of P.L. 103-66 was to target families at risk to prevent the need for foster care and to provide support to families in crisis. At the time the respondents completed the survey under report, agencies were providing family preservation and family support services funded in part by federal funds under the Social Security Act, Title IV-B, Subpart 2, and therefore special populations, including children of color, should be targeted for services.

"Despite significant advances in family preservation policy, practice, and programs, there was a rising chorus of criticism of the services that help keep children with their families and prevent entry into care, and of foster care practices that return children to their families when the family is unable to provide adequate care" (Downs, et al., 2000, p.310). Such thinking fueled passage of the Adoption and Safe Families Act of 1997, P.L. 105-89. With its passage, issues of child safety and permanence are currently at the fore of public policy implementation and practice consideration.

Special populations under P.L. 105-89 (ASFA). The data reported herein suggest that when legislative mandates required a programmatic focus on services intended to eliminate the need for foster care placement, giving special consideration to victims of foster care drift and special populations, most child welfare workers opted not to target services accordingly. Now that the legal standard required by P.L. 96-272 for reasonable efforts to keep children with their families is no longer unilaterally a stated policy goal (i.e., passage of P.L. 105-89, ASFA, results in circumstances for which reasonable efforts to keep children with their families are not required), can we be assured that all that can be done before children enter foster care will occur so that we can be comfortable with pursuing adoption for permanency?

Special populations under non-child welfare but related legislation. By virtue of the "risk" factors alone, many, if not all, of the previously explicated groups of special populations clearly receive protection under child welfare legislation. To strengthen an already compelling case, mental health as well as disabilities legislation also provides the rationale for why certain groups of children are classified as special populations.

Methodology

Design

This research project focused on an examination of the decision-making process involved in targeting children and families for intensive family preservation services. The issue of targeting was examined through an exploration of a critical approach to service delivery, namely, the practice of targeting special populations. The purpose of the research was to evaluate how family preservation workers handle the issue of special populations in deciding service delivery. "A Theoretical Model of Decision Making in Family Preservation Programs" guided inquiry (Denby, 1995). One major component of the decision making model involves the role played by workers in decisions to target service populations. The theory maintains that such factors as workers' values, biases, and characteristics have an effect on their attitudes, belief structure, and their subsequent behavior. Ultimately, these factors influence service delivery. The overall design for this study was exploratory-descriptive research that used the cross-sectional survey method. Specifically, the mail-survey method was used and was designed according to Dillman's (1978) "Total Design Method (TDM)."

Research Questions

Researchers explored the role played by workers in decision making, by posing the following questions: (1) are special populations targeted for family preservation services?; (2) if agencies do not use special populations as a service criterion, then what families are served?; (3) what factors hinder workers from using the special population service criterion?; and (4) why do those workers who favor using the special populations service criterion not use it?

Sample and Instrumentation

A list of 250 agencies was generated from the National Resource Center on Family-Based Services Annotated Directory of Selected Family-Based Services Programs, 1994, using a systematic random sampling technique. Each agency received two instruments which produced an attempted sample of 500 family preservation workers.

The study's research questions were pursued through the use of the "Decision-Making Survey," a 127-Likert item scale. In keeping with the literature which was discussed previously, the main measure, special populations, was operationalized for the respondents by providing a list of 16 types of children considered within child welfare to be a special population. The definition included the following: children of color, medically vulnerable, homeless children, sibling groups, children of the mentally ill, reunification families, potential adoption disruption families, older children, chronic juvenile offenders, severely emotionally disturbed (SED), children under the age of five, first-time parents, perinatal parents, children with birth abnormalities, children who are HIV positive or those who have AIDS, and status offenders. Although all sixteen types of special populations were studied, this article focuses on the findings which related to children of color. The instrument was developed as a probe for ascertaining the frequency and extent to which workers use the service criterion special populations as a target for service delivery.

Face, content, and construct validity. Survey items were developed using two sources: (1) results of a study on imminent risk (Walton & Denby, 1995); and (2) empirical literature. Themes that emerged from the Walton and Denby inductive analysis of 71 interviews with family preservation workers and administrators who represented both contract and referring agencies, were used to construct the "Decision-Making" instrument. The "Targeting Services by Special Populations," portion of the instrument included 52 questions. These questions pertain to four main constructs: (1) attitudes about the use of special populations as a targeting strategy; (2) the frequency of the use of the criterion special population in service delivery (behavior variable); (3) beliefs that influence the use or nonuse of special population; and (4) attributes associated with workers and the agencies for which they work. Additionally, the instrument was constructed in such a manner as to identify populations which are receiving service in the absence of the criterion special population.

Reliability. After close adherence to the TDM, a 60% return rate was achieved. This return rate is considered "good" for this level of survey research (Babbie, 1986). Nonetheless, Miller and Smith (1983) recommend a technique for assuring that nonresponse error is not a factor in conducting survey research. Miller and Smith instruct researchers to compare respondents to nonrespondents by taking a random sample (10-20%) of both, and comparing them on demographic data by computing a t test. The t test ascertains whether there is a difference between those who answer the survey and those who do not. A random sample of 20% of early and late respondents was selected for an analysis of nonresponse bias. Results of the t-test indicate categorically, that there is a non significant difference between the two groups' mean scores. There is no difference on demographic data between those respondents who returned their surveys early and those who returned them late. Therefore, because research has suggested that late respondents are most like nonrespondents, it can be inferred that those family preservation workers who did not return their surveys were no different from those who did; thus, nonresponse bias is assumed a non factor.

The "Decision Making" survey was pilot tested prior to implementation. Reliability scores on Part II, "Targeting Services by Special Populations," were strong. The Cronbach Alphas for the six sub-scales ranged from .63 to .98, suggesting strong inter-item correlation.

Various descriptive and inferential statistics were used for data analysis, including nonparametric tests and advanced multi-variate techniques. Specific to the findings reported in this article, the following tests were used for analyzing the data: descriptive statistics were used to organize the raw responses into indices that summarized the entire set of data; and correlation matrices were executed to ascertain whether a co-relationship exists among workers' willingness/nonwillingness to target services to "special populations" and moderating variables such as supports and barriers to targeting practices.

Findings

Demographics

The majority (63%) of the respondents in this study represent private, nonprofit agencies. Public agencies comprise the next highest category (31%). Most (72%) workers carry caseloads of 1-10 families. Twenty-four percent (24%) of the respondents reported that their cases can remain open for up to 10 weeks. Twenty percent (20%) of the workers service cases 11-15 weeks.

There is a wide range of age categories. More than half (56%) of the respondents are under the age of 40. A third of the respondents are 41-50 years of age. The remaining respondents are over the age of 50. The majority (76%) of the respondents are female. A third of the respondents report 1-5 years of experience in social services. Another third of the sample has 6-10 years of experience. The remaining third have anywhere from 11-20 years of experience. Overwhelmingly, the vast majority (65%) of the sample have worked in family preservation services for only 1-5 years. Twenty-three percent (23%) have 6-10 years of experience in family preservation.

One-half of the sample reported using a "family systems" approach to treatment. Twenty-two percent (22%) use a modified Homebuilders or some other approach. Nearly seventy percent (70%) of the sample's clients enter the service system primarily as a result of child abuse and neglect. Nearly eighteen percent (18%) report that their clients are largely referred as a result of mental health difficulties. The majority (70%) of the sample indicate that their primary service population is European American. African American and Hispanic clients only comprise twenty-one percent (21%) and three percent (3%), respectively, of the primary service population.

The majority (63%) of the respondents work in family preservation units that are housed within major agencies. Eighty-three percent (83%) of the sample are European American. Workers of color make up the remaining seventeen (17%). The majority (42%) of the sample possess a master's degree. Another twenty-two percent (22%) have at least a bachelor's degree. A near even split, forty-nine percent (49%) of the sample has social work degrees, while the other forty-six percent (46%) hold nonsocial work degrees.

Special Populations are not the Target of Family Preservation

Strong targeting support for any of the 16 groups of special population could not be found in this study. In fact, the greatest disagreement for targeting services to special populations occurred in the category of children of color (M = 2.35). Conversely, the group of children most strongly supported were children under the age of five (M = 2.85). The scores for the other groups of children are as follows: medically vulnerable (M = 2.65), homeless children (M = 2.78), sibling groups (M = 2.59), children of the mentally ill (M = 2.79), reunification families (M = 2.75), potential adoption disruption families (M = 2.73), older children (M = 2.61), chronic juvenile offenders (M = 2.48), severely emotionally disturbed--SED (M = 2.73), first-time parents (M = 2.65), perinatal parents (M = 2.50), children with birth abnormalities (M = 2.50), children who are HIV positive or those who have AIDS (M = 2.69), and status offenders (M = 2.51). (1)

Table 1 represents one of the sub-scales contained in the Special Population section of the "Decision-Making Survey." The questions sought workers' perceptions of their own behaviors and action with regard to the special population service criterion. Although items "A" and "B" may suggest that the respondents in this study do not exclude special populations, the remaining behavior items, collectively, suggest that workers do not strive to include special populations. Workers do not believe that family preservation services should be targeted using the service criterion, special population. For example, more than half (53%) (see Item "D") of the respondents disagree with establishing service eligibility policy that gives preference to special populations. Additionally, more than half of the workers (see Item "E") report that the majority of their most recent cases have not been special populations. Workers report that they have not advanced the notion that special populations ought to be the primary service criterion (Item F). Ninety-four percent (94%) of the sample indicated that they have not intervened with a referring source to the extent that they've requested that non-special populations not be the primary type of referrals they receive (see Item "G"). Finally, nearly the entire sample (93%) has not returned a referral to a referring agency because the referral did not fall under the eligibility criterion special population.

The Children and Families who do Receive Services

Workers were asked to define those categories of children and families who comprise the family preservation service population. Workers report that fifty percent (50%) of the clients who receive family preservation services represent the "general population" and do not possess any distinguishing features that would set them apart from other child welfare clientele. The other half of the family preservation population comprises twenty-five percent (25%) special populations, twelve percent (12%) "imminent risk," and another thirteen percent (13%) percent whose service criteria are unclear or "unknown."

Factors that Hinder Workers' Targeting of Special Populations

Table 2 is a display of workers' opinions of why they do not target family preservation services to special populations. In cases where a worker disagrees with the special populations service criterion, agency-level, macro-structural issues do not hinder him/her from targeting services to special populations. Likewise, researchers could not find overwhelming support for the premise that such moderating variables as agency policy or funding specifications cause workers to avoid using special population as the primary service criterion. Seemingly, what hinders workers' use of the service criterion special population, is the manner in which the term is conceptualized and the workers' own internal ideologies and/or biases. Workers expressed the following:

 the term special population does not have real significance
 because it has come to mean everyone (Item G, M = 3.4);
 so few cases are what they would actually consider
 special populations (Item B, M = 3.05);

 special populations demand more time and resources than
 what the workers feel they have available (Item D, M = 3.17); and

 workers felt that the lack of community resources available to aid
 them in their delivery of services to special populations
 (Item H, M = 3.39), could potentially be the hindering factor.

Several "belief structure" variables proved to be significant in attempts to understand why workers who are in favor of using the special population service criterion do not use it. When workers are in support of the special population service criterion but still do not use it to make a service decision, agency policies and the lack of community resources account for their actions. For example, moderate support (R = .48, p = .003) was found for the premise that agency-level barriers may influence workers who agree with the criterion but do not use it. Strong correlations (R = .64, p = .001) were found between workers' desires to target services to special populations and their perception that agency eligibility criteria are too general and thus impede their ability to target special populations. Finally, the problem of few community resources again surfaced as a factor imposing on the use of the criterion, special population. The community resource variable is not only positively correlated (R = .60, p = .000) with a favorable attitude toward the special population service criterion, but it is one of the few study variables which distinguishes between family preservation workers who have delivered service based on the criterion and those who have not.

Discussion

The results of this study indicate that special populations are not the target of intensive family preservation services. The results are consistent with a content analysis of the sampling frame used to draw the names of agencies. Recall, the Annotated Directory of Selected Family-Based Services Programs served as the major sampling frame for the study. A content analysis of the 368 programs registered with the Annotated Directory, revealed that 212 of the programs do not indicate within their program description, an official policy of targeting services to special populations. It seems that even in the face of legislative mandates and policy directives that emphasize service priority to special populations, workers do not utilize such classifications in their decision-making. In this study, there appears to be two parallel issues that are in operation. First, the study consisted of a significant number of workers who do not believe services should be targeted under the service criterion, special populations. Their belief structure is prompted by the manner in which they believe the term special population is defined. Seemingly, workers opposed to the special population service criterion also appear to be so because of some personal ideologies. Those workers that do not agree with targeting services to special populations are not hindered by agency-level, macro-structural issues. Second, in this study researchers discovered that there are a number of workers who do believe that special population ought to be used as a service criterion. Interestingly, unlike the workers who disagree with the special population criterion, workers who do agree, but do not target, are in fact hindered by agency-level, macro-structural issues. The only belief structure that is common to both groups of workers (i.e., those who support "special populations" and those who do not) is the opinion that there are too few community resources to aid the practice of targeting services to special populations.

Best Interpretation of the Results

Exploratory studies that rely heavily on descriptive and correlation analyses have inherent limitations. Given the limitations, results from this study should be interpreted cautiously. One potential limitation of the study concerns the issue of construct validity. There are three types of construct validity: convergent validity (measures correlate with the variables of the study in a predicted manner), discriminant validity (measures do not correlate with the variables that they should not), and sensitivity to change (the study's interventions produce expected change in observed measures and variables) (Anastasi, 1988 In Bloom, Fischer, & Orme, 1999). In this study the belief variables that measure workers' attitudes toward the criterion special population highly correlate in the predicted manner with the variables that measure workers' behavior; thus, there is evidence of convergent and discriminant validity. However, there is always the possibility that the manner in which the researchers operationally defined the notion special populations might not convey the same meaning for respondents. Although great care was given to operationalize the term special populations into categories of 16 groups, respondents may not have retained the specific definition as they answered other parts of the survey. Nonetheless, given the richness of the data, a solid direction for follow-up research is provided.

Implications and Conclusion

Strategies for Reforming Family Preservation Programs

The findings of this study have important implications for policy and practice. The field of intensive family preservation services is in an era in which its very efficacy is being called into question. Family preservation programs are currently scrutinized because program officials are unable to clearly identify who they believe family preservation services benefit most. The criticism of family preservation services abounds under a legislative cloud which prompts states to do something to reduce out-of-home placement rates. Moreover, legislative directives as far back as P. L. 96-272, P. L. 103-66, and now, P. L. 105-89, contain language which indicates that services are to address the needs of special populations. Taken together, all of these factors (i.e., program criticism, legislative mandates) prompt the question, "Is there a need to reform family preservation services?" The literature providing a rationale in support of targeting special populations, given their predicament within the child welfare system, is readily available and it is decisive. It would seem that if family preservation programs are to undergo restructuring, the "targeting dilemma" would be a good place to start. Deciding what families are most appropriate for services, is imperative. In any discourse concerning practice and policy reform, attention should be given to the plight of special populations. All decision-makers have a role in reforming intensive family preservation services and assuring that programs give attention to the needs of special populations.

Correcting the conceptualization of the term special population. State legislators can adopt specific targeting guidelines that define who special populations are within their jurisdiction. They can mandate committees to oversee the implementation of newly developed targeting guidelines. Additionally, workers have a vital role in the reform process. There is a need to emphasize who special populations are, and why they are. In this study, workers thought the term special population has come to mean virtually everyone. Workers can use practice wisdom to assist in defining who special populations are. Practice-inspired definitions should take into account such factors as general risk, legislation guides, and child vulnerability issues like age and ethnicity. As states work to shape policy directives that take into account the needs of special populations, child safety should always be the primary service goal no matter the classification of special population.

Combating bias ideologies. Additionally, referring agents might consider a practice of contracting exclusively with service agencies that utilize proven, culturally-specific intervention models. Finally, program managers share a critical role in reforming services and making sure programs address the needs of special populations. Program managers can begin to implement demonstration projects that target services to special populations.

The resource issue. Out-of-home placement among special populations like children of color, is a complicated issue and there certainly are no easy answers. Although the focus of this study was an examination of the attitudes, beliefs, and behaviors held by family preservation workers, program reform is by no means their sole responsibility. In fact, this study found that a significant number of workers do believe that the criterion special population should be used in service decisions. These workers are hindered in their efforts to target because there are not enough resources to support their practice. It seems that multi-level, system-wide changes are needed. If a wide-base of support for targeting special populations can be obtained among professionals, change should be desired and certain.

Table 1

Extent to Which Special Populations are Targeted for Service
Delivery

Item Frequency

a) Of the last eight cases, how many were served 0 94
 primarily on the basis that a child was considered 1-3 84
 to be a special population? 4-8 70
Total: 248

b) Within the last year, have you decided not to Yes 24
 open a case because it was a special population? No 226
Total: 250

c) Have you disagreed with someone because you Yes 39
 wanted to deliver services primarily to special No 212
 populations?
Total: 251

d) If you had to establish a policy for intensive Yes 137
 family preservation service eligibility, would you No 107
 eliminate rules which state that special
 populations will be given priority consideration?
Total: 244

e) Have the majority of your most recent cases been Yes 113
 special population types? No 134
Total: 247

f) Have you felt that special populations should be Yes 25
 the primary criterion by which services are No 225
 delivered and then told someone?
Total: 250

g) Have you requested that a referring source stop Yes 5
 sending cases that are not special populations? No 241
Total: 246

h) In recent months, I have returned a referral Yes 9
 because it was not a special population case. No 237
Total: 246

Item Percent

a) Of the last eight cases, how many were served 36.7%
 primarily on the basis that a child was considered 32.8%
 to be a special population? 27.3%
Total: 96.8%

b) Within the last year, have you decided not to 9.4%
 open a case because it was a special population? 88.3%
Total: 97.7%

c) Have you disagreed with someone because you 15.2%
 wanted to deliver services primarily to special 82.8%
 populations?
Total: 98.0%

d) If you had to establish a policy for intensive 53.5%
 family preservation service eligibility, would you 41.8%
 eliminate rules which state that special
 populations will be given priority consideration?
Total: 95.3%

e) Have the majority of your most recent cases been 44.1%
 special population types? 52.3%
Total: 96.4%

f) Have you felt that special populations should be 9.8%
 the primary criterion by which services are 87.9%
 delivered and then told someone?
Total: 97.7%

g) Have you requested that a referring source stop 2.0%
 sending cases that are not special populations? 94.1%
Total: 96.1%

h) In recent months, I have returned a referral 3.5%
 because it was not a special population case. 92.6%
Total: 96.1%

Note: Percent scores which sum to less than 100 contain missing data.

Table 2

Factors That Hinder Workers' Use of Special Population as a Service
Criterion

 Standard
Item Mean Deviation n

a) Although I support the notion 1.91 0.562 183
 that services should be targeted
 to special populations, my
 agency's policies do not allow
 such a practice.
b) Although I support the notion 3.05 0.466 199
 that services should be targeted
 to special populations, so few
 cases are what I would consider
 special populations.
c) Although I support the notion 2.1 0.631 197
 that services should be targeted
 to special populations,
 referring or funding sources
 discourage such a practice.
d) Although I support the notion 3.17 0.59 200
 that services should be targeted
 to special populations, these
 types of cases demand more time
 and resources than we have.
e) Although I support the notion 2.08 0.621 182
 that services should be targeted
 to special populations, the
 agency's elibility criteria are
 so general, they hinder the use
 of special populations as a
 service criterion.
f) Although I support the notion 2.06 0.602 192
 that services should be targeted
 to special populations, the
 agency's treatment model is not
 conductive to directing services
 specifically to special
 populations.
g) Although I support the notion 3.4 0.544 205
 that services should be
 targeted to special populations,
 the term "special populations"
 has come to mean everyone.
h) Although I support the notion 3.39 0.525 203
 that services should be targeted
 to special populations, there
 are few community resources to
 help us deliver services to
 these groups.

Note: Scale values range from 1 to 4, where 1 = Strongly
Disagree, 2 = Disagree, 3 = Agree, and 4 = Strongly Agree.

Note

(1.) Scale values for the subscale "Attitude Toward Using Special Population as a Service Criterion" range from 1 to 4, where 1 = Strongly Disagree, 2 = Disagree, 3 = Agree, and 4 = Strongly Agree.

References

Administration for Children and Families--ACF (2002). Fact Sheet--Protecting the Well-Being of Children. http://www.acf.dhhs.gov/programs/opa/facts/ chilwelfpr.htm, 2/12/02.

Allen, M., Golubock, C., & Olson, L. (1983). A guide to the Adoption Assistance and Child Welfare Act of 1980. In M. Hardin (Ed.), Foster children in the courts (pp. 575-600). Boston, MA: Butterworth Legal Publishers.

Anderson, V. (1998). HIV infection and children: A medical overview. Child Welfare, 77(2), 107-114.

Babbie, E. (1986). The practice of social research (4th ed.) Belmont, CA: Wadsworth Publishing Co.

Berliner, L. (1993). Is family preservation in the best interest of children? Journal of Interpersonal Violence, 8(4), 556-557.

Berrick, J. D., Needell, B., Barth, R. P, & Jonson-Reid, M. (1998). The tender years: Toward developmentally sensitive child welfare services for very young children. New York, NY: Oxford University Press.

Berthier, M., Oriot, D., Bonneau, D., Chevrel, J., & et al. (1993). Failure to prevent physical child abuse despite detection of risk factors at birth and social work follow-up. Child Abuse and Neglect, 17(5), 691-692.

Black Administrators in Child Welfare--Annual Symposium (2001). The state of African American children 2001 fact sheet. Empowering Leaders in Child Welfare: Champions for African American Children, Families, and Communities, March 5-6, 2001.

Bloom, M., Fisher, J., & Orme, J. G. (1999). Evaluating practice: Guidelines for the accountable professional. Boston, MA: Allyn and Bacon.

Cameron, T. (2000). Proposed initiatives for healthy children orphaned by AIDS. Journal of Health and Social Policy, 11(4), 15-39.

Coiro, M. J. (1998). Maternal depressive symptoms as a risk factor for the development of children in poverty. Dissertation abstracts international: Section B: The sciences & engineering, 58(7-B), 3943.

Courtney, M. E. (1994). Factors associated with the reunification of foster children with their families. Social Service Review, 68, 81-108.

Denby, R. W. (1995). Targeting families for family preservation services: The decision-making process. Unpublished doctoral dissertation. The Ohio State University. Columbus, Ohio.

Dillman, D. A. (1978). Mail and telephone surveys: The total design method. New York, NY: John Wiley.

Douglass, A. (1996). Rethinking the effects of homelessness on children: Resiliency and competency. Child Welfare, 75(6), 741-751.

Downs, S., Moore, E., McFadden, E., Costin, L. (2000). Child welfare and family services: Policies and practice, 6th edition. Boston: Allyn and Bacon.

Draimin, B. H., Gamble, I., Shire, A., & Hudis, J. (1998). Improving permanency planning in families with HIV disease. Child Welfare, 77(2), 180-194.

Drapeau, S., Simard, M., Beaudry, M., & Charbonneau, C. (2000). Siblings in family transition. Family Relations, 49(1), 77-85.

Feldman, L. (1990). Target population definition. In Y. Yuan & M. Rivest, (Eds.), Preserving families (pp. 16-38). Newbury Park: Sage.

Finzi, R., & Stange, D. (1997). Short term group intervention as a means of improving the adjustment of children of mentally ill parents. Social Work With Groups, 20(4), 69-80.

Frame, L., Berrick, J. D., & Brodowski, M. L. (2000). Understanding reentry to out-of-home care for reunified infants. Child Welfare, 79(4), 339-369.

Goicoechea-Balbona, A. (1998). Children with HIV/AIDS and their families: A successful social work intervention based on the Culturally Specific Health Care Model. Health and Social Work, 23(1), 61-69.

Gustavasson, N.S., & Segal, E.A. (1994). Critical issues in child welfare. Thousand Oaks, CA: Sage Publications.

Heneghan, A. M., Horwitz, S. M., & Leventhal, H. M. (1996). Evaluating intensive family preservation programs: A methodologic review. Pediatrics, 97, 535-542.

Jonson-Reid, M., Williams, J. H., & Webster, D. (2001). Severe emotional disturbance and violent offending among incarcerated adolescents. Social Work Research, 25(4), 213-222.

Kamerman, S., & Kahn, A. (1993). If child protective service is driving child welfare--Where do we go from here? Public Welfare, 51, (1), 41-43.

Kinney, J., Haapala, D., Booth, C., & Leavitt, S. (1991). Keeping families together: The Homebuilders Model. New York: Aldine De Gruyter.

Kirk, R. S. (2000). Final report: Retrospective evaluation of intensive family preservation services. Chapel Hill North Carolina: Jordan Institute for Families.

Loeber, R., & Farrington, D. P. (2000). Young children who commit crime: Epidemiology, developmental origins, risk factors, early interventions, and policy implications. Development & Psychopathology, 12(4), 737-762.

Luntz, J. (1995). The invisible dimension in mental health: Children whose parents suffer from a serious mental illness. Australian Social Work, 48(3), 19-27.

Maluccio, A., Pine, B. A., & Warsh, R. (1994). Protecting children by preserving their families. Children and Youth Services Review, 16(5/6), 295-301.

Mason, S. (1998). Custody planning with HIV-affected families: Considerations for children welfare workers. Child Welfare, 77(2), 161-177.

Miller, L. E., & Smith, K. L. (1983). Handling nonresponse issues. Journal of Extension, 21, 45-50.

National Association of Black Social Workers (1992). Preserving African-American families: Research and action beyond the rhetoric. Detroit, MI: Author.

National Association of Social Workers (NASW). School counseling, stable families measures pass. NASW News, (February, 2002).

National Black Child Development Institute. (1989). Who will care when parents can't? A study of Black children in foster care. Washington, DC: Author.

Nelson, K. (1994). Do services to preserve the family place children at unnecessary risk? No. In E. Gambrill & T. J. Stein (Eds.), Controversial issues in child welfare (pp. 67-71). Needham Heights, MA: Allyn & Bacon.

Neuspiel, D. R., Zingman, T. M. (1993). Custody of cocaine-exposed newborns: Determinants of discharge decisions. American Journal of Public Health, 83, 1726-1729.

Nugent, W. R., Carpenter, D., & Parks, J. (1993). A statewide evaluation of family preservation and family reunification services. Research on Social Work Practice, 3(1), 40-65.

Pecora, P. J. (1994). Are intensive family preservation services effective? Yes. In E. Gambril and T. J. Stein (Eds.), Controversial issues in child welfare (pp. 290-303). Needham Heights, MA: Allyn & Bacon.

Pinderhughes, E. E. (1991). The delivery of child welfare services to African American clients. American Journal of Orthopsychiatry, 61(4), 599-605.

Roberts, D. (2002). Shattered bonds: The color of child welfare. New York, NY: Basic Civitas Books.

Rznepniki, T. L. (1994). Are intensive family preservation services effective? No. In E. Gambril and T. J. Stein (Eds.), Controversial issues in child welfare (pp. 303-309). Needham Heights, MA: Allyn & Bacon.

Scherer, D. G., Brondino, M. J., Henggeler, S. W., Melton, G. B., et al. (1994). Multisystemic family preservation therapy: Preliminary findings from a study of rural and minority serious adolescent offenders. Journal of Emotional & Behavioral Disorders, 2(4), 198-206.

Schuerman, J. R., Rzepnicki, T. L., & Littell, J. H. (1994). Putting families first: An experiment in family preservation. New York, NY: Aldine De Gruyter.

Seader, M. B. (1994). Do services to preserve the family place children at unnecessary risk? Yes. In E. Gambrill & T. J. Stein (Eds.), Controversial issues in child welfare (pp.59-65, 71-72). Needham Heights, MA: Allyn & Bacon.

Sedlak, A. J., (1997). Risk factors for the occurrence of child abuse and neglect. Journal of Aggression Maltreatment & Trauma, 1(1), 149-187.

Smith, M. C. (1996). An exploratory survey of foster mother and caseworker attitudes about sibling placement. Child Welfare, 75(4), 357-375.

Smith, N. A. (2000). Understanding the lack of family reunification success for chemically dependent mothers and their children: A presentation of consumer and service provider perspectives. Dissertation abstracts international, A: The humanities and social sciences, 60(10), 3798A.

Spar, K. (1993) The family preservation and support program: Background and description. Congressional Research Service, The Library of Congress. Washington, DC.

Stehno, S. (1990). The elusive continuum of child welfare services: implications for minority children and youths. Child Welfare, 69(6), 551-562.

Stevenson, K., Cheung, K., & Leung, P. (1992). A new approach to training child protective services workers for ethnically sensitive practice. Child Welfare, 71(4), 291-305.

Solnit, A. J., Adnopoz, J., Saxe, L., Gardner, J., & Fallon, T. (1997). Evaluating systems of care for children: Utility of the clinical case conference. American Journal of Orthopsychiatry, 67(4), 554-567.

Talbot, E. B. P. (2001). Successful family reunification: The contribution of clinical social work. Dissertation abstracts international, A: The humanities and social sciences, 62(1-A), 335.

Taylor-Brown, S., Teeter, J. A., Blackburn, E., Oinen, L., & Wedderburn, L. (1998). Parental loss due to HIV: Caring for children as a community issue. Child Welfare, 77(2), 137-160.

Tenner, A., Feudo, R., & Woods, E. R. (1998). Shared experiences: Three programs serving HIV-positive youths. Child Welfare, 77(2), 222-250.

Triseliotis, J. (2002). Long-term foster care or adoption? The evidence examined. Child and Family Social Work, 7(1), 23-36.

U.S. Code Congressional and Administrative News. 103rd Congress, First Session, 1993, Vol. 1., Part B, Child and Family Services, Subpart 2, Family Preservation and Support Services, pp. 649-658. St. Paul, Minnesota: West Publishing Company.

U.S. Code Congressional and Administrative News. 105th Congress, First Session, 1997 Vol. 4, Adoption and Safe Families Act of 1997, PL 105-89. House Report #105-77. St. Paul, Minnesota: West Publishing Company.

U.S. Code Congressional and Administrative News. 96th Congress, 2nd Session, 1980, Vol. 3, Adoption Assistance and Child Welfare Act of 1980, PL 96-272. St. Paul, Minnesota: West Publishing Company.

U.S. Department of Health and Human Services (2001). Evaluation of family preservation and reunification programs: Executive summary to the interim report. Washington, DC: U.S. Government Printing Office.

U.S. Department of Health and Human Services, Children's Bureau (1997). National study of protective, preventive, and reunification services delivered to children and their families. Washington, DC: U.S. Government Printing Office.

U.S. Department of Health and Human Services. (1996). Federal Register, HHS consolidate planning requirements under subparts 1 and 2 of Title IV-B, CAPTA, and the independent living program under Title IV-E, Social Security Act.

W.K. Kellogg Foundation. (1995). Families for kids of color: A special report on challenges and opportunities. Battle Creek, MI: Author.

Walton, E., & Denby, R. W. (1997). Targeting families to receive intensive family preservation services: Assessing the use of imminent risk of placement as a service criterion. Journal of Family Preservation, 2(2), 53-70.

Walton, E., Fraser, M., Lewis, R., Pecora, P., & Walton, W. (1993). In-home family-focused reunification: An experimental study. Child Welfare, 72, 473-487.

Wang, C., & Daro, D. (1997). Current trends in child abuse reporting and fatalities: The results of the 1996 fifty state survey. National Committee to Prevent Child Abuse. Chicago, IL.

RAMONA W. DENBY, PH.D.
University of Nevada Las Vegas
School of Social Work

CARLA M. CURTIS, PH.D.
The Ohio State University
College of Social Work