Strengthening the Child Welfare Workforce: Lessons from Litigation

Abstract

The recruitment, preparation, support, and retention of public and private agency childwelfare staff working with abused and neglected children and their families are important and ongoing concerns. During the past two decades, many questions have been raised about the adequacy of the child welfare workforce and the supports provided to it. This article provides the findings from a review of efforts to strengthen the child welfare workforce in the context of class-action litigation for system reform. The lessons learned provide a useful framework for current and future efforts to improve the child welfare workforce, both within and without the context of litigation.

KEYWORDS:

• child welfare
• workforce
• litigation
• systemic change

Introduction

The recruitment, preparation, support, and retention of public and private child welfare staff working with abused and neglected children and their families are important and ongoing concerns in the child welfare field (Alliance for Children and Families [ACF], 2004; Annie E. Casey Foundation [AECF], 2003; Child Protection Report, 2002; Institute for the Advancement of Social Work Research [IASWR], 2005). Over the past few decades, many questions have been raised about the adequacy of preparation and support provided to the child welfare workforce (ACF, American Public Human Services Association [APHSA], & Child Welfare League of America [CWLA], 2001; Alwon & Reitz, 2000a; 2000b; Ellett, 2002; Mor Barak, Nissly, & Levin, 2001; Pelton, 1990), as research has highlighted the impact of workforce issues on outcomes for children, youth, and families, and on expenditures at the federal, state, and local levels.

Caseworkers with high caseloads have less time to interact with children and families, develop and implement thoughtful case plans and provide appropriate supervision of placements (Hess, Folaron, & Jefferson, 1992; Stein, Gambrill, & Wiltse, 1978, cited in Stein & Callaghan, 1990; United States General Accounting Office [US GAO], 2003), factors that have been found to contribute to the reentry of children into foster care (Hess, Follaron, Jefferson & Kinnear, 1991; Hess et al., 1992). The United States Department of Labor estimates that the cost of worker turnover is approximately one-third of a caseworker's annual salary, meaning child welfare agencies ‘take a hit’ of approximately $10,000 every time a worker leaves (Flower, McDonald, & Sumski, 2005; Graef & Hill, 2000). The quality of the child welfare workforce becomes a concern of the general public—if only momentarily—when the media reports on the deaths of children previously or currently known to the child welfare system, and poor casework appears to have contributed to the deaths.

Class-action litigation for child welfare has focused on reforming public child welfare systems across the country (Bertelli, 2004; CWLA & the ABA Center on Children and the Law, 2005; Kaplan, 2003). By creating a greater awareness of systemic problems, putting pressure on politicians and administrators, and using the court to promote change, litigation has served as a catalyst for child welfare reform (Center for the Study of Social Policy [CSSP], 1998; Gluck Mezey, 1998) by enhancing resources and addressing critical areas, including caseloads, staff qualifications, training, supervision, recruitment and retention, policy development, information systems, quality assurance, and financing (CWLA & the ABA Center on Children and the Law, 2005). Limitations of class-action litigation as a reform tool that have been noted are its inherently adversarial nature and potential to focus more on indicators of quantity (bean-counting), and less on creating a comprehensive, strategic process focused on quality and deep institutional change (CSSP, 1998; Gluck Mezey, 1998).

Settlement agreements, consent decrees and other court ordered remedies (hereafter, court orders ¹ ) resulting from class action litigation in multiple jurisdictions across the nation have included provisions specifically targeted to improving the child welfare workforce (CWLA & the ABA Center on Children and the Law, 2005). When the courts mandate workforce improvements, they are legally enforceable and subject to ongoing monitoring. The implementation of a child welfare court order can therefore provide an opportunity to evaluate strategies to improve the workforce.

This article provides the findings from a review of efforts to strengthen the child welfare workforce in the context of class action system reform litigation. The lessons learned provide a useful framework for current and future efforts to improve the child welfare workforce, both in and outside of the context of litigation.

METHODOLOGY

Children's Rights (New York, NY), in collaboration with the National Center for Youth Law (NCYL, Oakland, CA) and with support from Cornerstones for Kids (Houston, TX), undertook an assessment of the workforce issues that precipitated and/or became the focus of class action litigation in 12 jurisdictions; the requirements (provisions) related to workforce issues that are included in the court orders in these cases; the factors associated with successful implementation of these requirements; and the barriers that have hindered success. The primary method of data collection was in-depth interviews with stakeholders, supplemented by a review of published reports documenting progress in these lawsuits.

Study Sample

Based on a review of the child welfare literature pertaining to workforce issues, Children's Rights and NCYL developed a definition of “workforce provision” in order to identify workforce provisions in court orders and select cases for inclusion in the project:

A workforce provision is any goal, benchmark, activity or intended outcome in a court order that directly targets developing and improving the quality and capacity of the child welfare workforce. This includes any type of provision that is designed to make the workforce better able to perform its duties and responsibilities; professionalizes the workforce; aids in the recruitment and retention of workers; improves the organizational environment and culture; or aims to develop improved systems of accountability, oversight, and evaluation of practice. The “workforce” includes all public and private-agency child welfare workers, supervisors and administrators (child protective services, intake, ongoing, foster care, adoption, etc.).

In 2005, Children's Rights and NCYL identified 18 major current and recent child welfare system reform class action lawsuits that potentially addressed workforce-related issues were initially identified. Following a review of these 18 lawsuits, six cases that were in active litigation or early settlement phase or did not specifically address any workforce-related issues were excluded. The 12 cases ultimately included in this study are listed in Table 1.

A total of 90 individuals associated with the 12 cases selected were identified as possible interviewees. Plaintiffs' counsel in these cases initially identified 50 individuals to be interviewed. During their interviews, these interviewees subsequently identified 40 additional potential interviewees. It should be noted that some interviewees held more than one role during and/or since the litigation, (such as interviewees who were public employees at one point and later became private-agency employees or advocates.) The overall response rate from the 90 potential interviewees was 82%. Of the prospective interviewees 10% did not consent to be interviewed. The remaining 8% of prospective interviewees were unable to be located or did not respond to requests to be interviewed.

The final sample of 74 interviewees included 20 plaintiffs' counsel, 21 current or former public child welfare agency representatives (includes public child welfare agency commissioners and staff, both current and former [i.e., not only parties who were specifically named as defendants in the lawsuit]), 11 current/former court monitors, 12 advocates (next friends; child advocates; foster/adoptive parent association representatives; service provider membership organization representatives; citizen review board members or other stakeholder monitoring group representatives), and 10 other individuals (juvenile court judges; university-agency training partnership staff; private providers; consultants; mediators). Most cases (nine of 12) had between five and seven interviewees. One case had three interviewees, and two cases had nine and 10 interviewees, respectively. In three of 12 cases, no current or former child welfare agency representatives were interviewed due to agency counsel declining to grant consent. In these three cases (as in all of the cases), advocates, monitors, and private providers were among those interviewed, in addition to plaintiffs' counsel.

TABLE 1 Child Welfare Class Action Lawsuits Included in this Review

Case name	Jurisdiction	Date of initial complaint filing	Date of initial court order	Date case ended
Angela R.	Arkansas	July 1991	April 1992	December 2001
B. H.	Illinois	June 1988	December 1991	N/A
Brian A.	Tennessee	June 2000	July 2001	N/A
David C.	Utah	February 1993	May 1994	N/A
G. L.	Kansas City, MO	March 1977	March 1983	February 2006
Jeanine B.	Milwaukee, WI	June 1993	December 2002	N/A
Joseph A.	New Mexico	July 1980	September 1983	February 2005
Juan F.	Connecticut	December 1989	January 1991	N/A
L. J.	Baltimore, MD	December 1984	September 1988	N/A
LaShawn A.	Washington, DC	June 1989	November 1993	N/A
R. C.	Alabama	November 1988	June 1991	N/A
Ward	Broward County, FL	October 1998	May 2000	October 2001

Measures

A structured interview protocol was developed by Children's Rights and NCYL. The interview protocol included dichotomous, multiple-choice, and scaling questions, as well as open-ended questions. The protocol was designed to elicit responses to the following general questions:

	What issues precipitated the lawsuit? How were the issues precipitating the lawsuit identified? Were workforce issues specifically identified?
	How longstanding were the identified issues?
	How appropriate or effective were the initial provisions in the court orders? How appropriate/effective are the current provisions?
	Is anything particularly unique to the case in jurisdiction?
	What is the current status of the child welfare workforce in relation to the provisions of the court order/settlement agreement in the following workforce-related areas: recruitment and retention; caseload and workload; supervision; education; training; quality assurance; organizational environment; employment incentives; data; research and evaluation; technological resources; workspace; funding and legislative support?
	Has progress been made on all, some, or none of these provisions? In what areas has there been substantial or moderate improvement? In what areas has there been little or no improvement?
	What strategies have been implemented?
	What have the barriers to improvement been?
	What steps should or should not have been taken to promote greater improvements?
	Would improvements have occurred without the court orders? Why or why not?

To supplement the information provided by interviewees, Children's Rights and NCYL staff gathered initial complaints; ² monitoring reports; ³ and other documentation containing data reflecting the progress made on workforce issues in these jurisdictions, in order to collect data in the following workforce-related areas: recruitment and retention; caseload and workload; supervision; education; training; quality assurance; organizational environment; employment incentives; data; research and evaluation; technological resources; workspace; funding and legislative support.

Data-Collection Procedures

All interviewees were promised confidentiality, i.e., that their specific comments would not be attributed to them in the final report. Current and former child welfare agency representatives whose agency counsel permitted their participation were also promised that the information they provided during their interviews could not be used in the context of current or future litigation.

Telephone interviews were conducted from August 2005 to May 2006 by Children's Rights and NCYL staff, including four attorneys and two policy analysts. These interviews were not recorded. In addition to collecting interviewees' specific answers to quantitative questions, extensive notes were taken regarding interviewees' comments and feedback on open-ended questions, with the assistance of social work or legal interns who acted as transcribers. All staff and interns involved in the project were provided with a 90-minute training on the structured interview protocol and data collection techniques prior to conducting or taking notes during interviews.

During September and October of 2006, Children's Rights and NCYL staff (including attorneys, paralegals, and policy analysts) reviewed initial complaints, monitoring reports and other documentation containing data reflecting the progress made on those selected workforce issues in the 12 jurisdictions.

Data Analysis

Quantitative data collected during interviews (e.g., role in the litigation, responses to Likert and other scaling questions, etc) were entered into an SPSS database (Version 14, SPSS, Chicago, IL) and analyzed. All of the additional commentary provided by interviewees was transcribed and reviewed. This information was collated into an outline organized by the workforce topics noted previously. These documents were then assessed by Children's Rights policy analyst staff to identify themes.

Children's Rights and NCYL also reviewed the quantitative data provided in litigation-related documents (e.g., complaints, monitoring reports) and summarized the progress identified on the range of workforce issues in these jurisdictions.

FINDINGS

Workforce Provisions in Court Orders

Children's Rights and NCYL reviewed the court orders in the 12 identified class-action cases for child welfare reform and categorized all of the provisions in these orders related to developing and improving the quality and capacity of the child welfare workforce. A total of 342 workforce provisions were identified in the court orders across the 12 cases. Some court orders contained a general provision on an issue, whereas others contained several very detailed individual provisions on elements of the same issue. Table 2provides examples of workforce provisions in some of the court orders reviewed.

Interviewees' Perspectives

Interviewees provided in-depth commentary on the steps that have been taken in their jurisdictions to improve the workforce, some of which were required by explicit provisions in the court order, and others, although not explicitly included in the court order, grew from the order and/or occurred in the context of litigation. Interviewees were extremely candid in their interviews, providing very rich and deep perspectives. The purpose of this review was to compile and take stock of the experiences and lessons learned in these jurisdictions so that this information can be shared and utilized to inform current and future efforts to improve child welfare workforce. There was a remarkable degree of consistency in the comments and recommendations across the various roles of the stakeholders, for example, the plaintiffs' counsel, child welfare agency representatives, and advocates. Their comments and recommendations provide a framework for improving the child welfare workforce in the context of litigation and other reform efforts.

ROLE OF DIFFERENT FACTORS IN IMPROVING THE CHILD WELFARE WORKFORCE

Interviewees were asked to rate the importance of various contextual factors in improving the child welfare workforce in their respective jurisdictions. Three factors were consistently identified by the majority of interviewees (more than 70%) to be the most important: 1) the leadership of the child welfare agency; 2) the resources, attention, and pressure resulting from the litigation; and 3) budget considerations. Another important factor identified by interviewees was support of or resistance to the reform effort by social workers and supervisors. Interviewees frequently noted the importance of a confluence of factors (e.g., the lawsuit brought critical attention, and strong agency leadership was necessary to implement the reforms).

Of all interviewees, 94% said that child welfare agency leadership was “very important” (78%) or “important” (16%) to improving the child welfare workforce. An additional 6% said that agency leadership was “somewhat important” for improving the workforce. As one interviewee (a training consultant) explained:

[The Commissioner] went out and worked with people and got close to the workforce, not in a micromanagement way, but gained their loyalty and support. He was very hands on, went on investigations with staff, tried to see the job through workers' eyes, and he really gained the support of the workers. At the same time, he wasn't trying to win the support of every single person, just a critical mass, so he could start to work with them to change the system.

TABLE 2 Sample Workforce Provisions Included in Court Orders

Jurisdiction	Type of provision	Provision
Tennessee	Recruitment and retention	In any region where the turnover rate exceeds 10% and in which cases are being left uncovered or reassigned to workers already at the caseload cap, the agency shall hire and maintain a pool of trained workers on stand-by.
Arkansas	Staff incentives	DHS shall enhance DCFS's ability to attract qualified applicants through its educational stipend program for undergraduate and graduate social work students, its request for salary upgrades for direct service staff, the capacity to allow direct service staff to work “flex time,” and the expansion of acceptable degrees for the position as family-service worker.
Utah	Supervision	If a supervisor identifies a problem based on review of a caseworker's compliance with various requirements and standards, he/she may take corrective action, including additional training, assigning a mentor, and/or having a supervisor accompany the worker during investigations. Workers who fail to improve their performance shall be subject to disciplinary action.
Utah	Training and education	Once all DCFS staff members have been trained to the model, a training and support group will be established in each region for front-line supervisors. The groups will be facilitated by a variety of people. The purpose of the support group is to discuss Practice Model implementation and recommended improvements. The group will also receive ongoing training with the expectation that the supervisor is to train his/her staff in the months between trainings. This continuous learning process will begin with training the supervisors in solution-based practice, with movement to other significant skills training each year. Ongoing monthly group for supervisors and mentors on how to coach staff on obtaining skills in direct practice and how to test for skill level attainment and retention.
Milwaukee, WI	Caseload and workload ^1	For each case management site: maximum average of 11 families/case-carrying manager (measured by averaging each site's current monthly caseload average with the corresponding site averages for the preceding 2 months).
Alabama	Organizational environment	The Child Welfare Consultant's function is to support the county around developing and stabilizing an infrastructure and maintaining a child welfare system that supports best practice. It also includes monitoring the system for best outcomes and identifying trends in an effort to proactively sustain the system of care.
Jackson County, MO	Accountability and oversight	Information from Quality Assurance Reviews must be reflected in Department training: Workers or supervisors for whom a quality review identifies problems must be referred for specialized training and supports. The Training Unit should keep data on those workers and supervisors referred for specialized training and follow-up to ensure that the requisite training has been provided and completed within 3 months of referral. Aggregate information from quality reviews must be transmitted on quarterly basis to the Training Unit for the development of training initiatives to remedy problems identified through quality reviews.
Alabama	Data	Each county shall report 3 months of caseload data by worker and a quarterly average on the quality assurance reports.
Washington, DC	Research and evaluation	The Department must formally assess (and, if necessary, modify) its placement support service programs at least once every 2 years.
Broward County, FL	Funding and legislative support	The Secretary has acted to reallocate the base-operating budget and new funds appropriated to better achieve staff equity among the 15 districts.

Of all interviewees, 93% indicated that the lawsuit and its resulting court orders and monitoring were “very important” (74%) or “important” (19%) to improving the child welfare workforce. An additional 7% said the lawsuit was “somewhat important” for improving the workforce. No interviewees said the lawsuit was “not important.” Of all interviewees, 87% said that there would not have been improvements without the lawsuit. An additional 11% said the improvements would not have been as significant or occurred as quickly without the lawsuit. According to one court monitor, “Folks would have never defined the workforce as a real problem worthy of fixing without the consent decree. These issues often require additional resources, and you couldn't deal with the problem without the pressure and spotlight of the consent decree.”

Of all interviewees, 92% said budget considerations were “very important” (74%) or “important” (18%) to improving the child welfare workforce. Also, 8% of interviewees (six interviewees representing six different jurisdictions) said budget considerations were only “somewhat important” or “not important” at all.

Of all interviewees, 73% indicated that social workers' and supervisors' support of various reform efforts and initiatives were “very important” (45%) or “important” (28%) to improving the child welfare workforce, and 12% said these were “somewhat important.” Also, 15% (11 interviewees representing seven jurisdictions) said this factor was “not important.” One child welfare agency representative noted, “Caseworkers and supervisors viewed the settlement agreement as their salvation.” A Children's Court Judge indicated that “Line social workers have bought into efforts to improve the child welfare system.” In contrast, another child welfare agency representative in a different jurisdiction explained “A lot of us agency staff looked at ourselves as victims in the beginning [of the lawsuit].”

COMMENTARY ON SELECT WORKFORCE ISSUES

Interviewees also offered their perspectives on a number of critical workforce issues.

Staff recruitment and retention

Most interviewees indicated that the litigation resulted in increased recruitment efforts, additional staff, and greater systemic attention to both recruitment and retention. However, interviewees said that more attention needed to be paid to retention, as gains made through successful recruitment can be quickly thwarted by high turnover. As one child welfare agency director noted:

You may be able to create greater capacity with a consent decree, as it has the promise of improvement. On the other hand, turnover doesn't necessarily get resolved. You may hire 1,000 workers to comply, but then you are losing folks even as you're hiring. It's about supporting workers once they're on board. You have to pay attention to the simple things (e.g., do they have a desk, a place to work) and ask “How do you prepare the workforce and support them?”

Benefits and compensation

Although staff in a few jurisdictions experienced gains in pay, most interviewees noted that worker salaries and benefits were unaddressed in many of the reform efforts. As one plaintiffs' counsel noted, “One of the things we could have pushed is to sustain the increases in salary and benefits. Creating a better salary and benefit structure would have helped. The more people you retain, the less money you have to spend training new staff.” Acknowledging the complexity of making change given state personnel systems and union contracts, interviewees spoke about the need to identify and attempt to rectify low salaries and undefined career paths, and to address salary and benefit disparities between the public and private sectors.

Supervision

Most interviewees noted that, since the litigation, supervisory ratios have improved, the amount of time caseworkers spend with their supervisors has increased and performance evaluations are completed in a more timely fashion. In several instances, however, interviewees said that the quality of supervision had not significantly improved. As a monitor in one case explained, “Although we addressed the number of supervisors, we didn't really address the quality of supervision. We had good requirements for education, training and experience for supervisors, but we didn't do enough exploration of what makes a good supervisor.”

Training

Almost all of the court orders in the 12 jurisdictions reviewed included specific requirements regarding number of training hours. Most interviewees noted that the litigation led to substantially increased training and the establishment of both educational requirements and opportunities for staff. As one representative of a child welfare agency expressed,

The agency training program is much improved. Formerly, new caseworkers were immediately given a full caseload and would be given some half-hearted training later on when they needed a break from the rigors of their caseload. Now, workers are given several weeks of training before they pick up a full caseload, and receive 40 hours per year of ongoing professional development training thereafter. Both initial and ongoing trainings are aimed at improving job performance and integrated into the agency's social work model and organizational goals.

Some interviewees commented that improvements in training and education had resulted in better casework practice. However, other interviewees said that outcomes had not yet been substantially affected by changes in training and education, and noted that there was a need to focus more on the quality of training.

Caseload

All of the court orders in the 12 jurisdictions reviewed included specific requirements regarding caseloads. Almost all of the interviewees said that the litigation played a substantial role in defining appropriate caseload limits and decreasing worker caseloads. They also noted, however, that progress has been hindered by worker turnover, the causes of which can include inadequate clerical support. While some interviewees said that clerical supports improved following the litigation, others indicated that needed administrative supports were largely overlooked during the reform process. As one advocate noted:

Caseloads are taken more seriously now, as they have hired more staff and reduced the caseloads from 30–50, to about 20 now. But they still don't get enough clerical or support staff, there are really no secretaries, and no case aides to speak of. So caseworkers don't get much direct support in their day-to-day responsibilities.

Working conditions

A few interviewees said that, following the court order, working conditions were substantially improved. These interviewees noted that office environments were cleaned and upgraded, desks and telephones became available for all workers, and safety protocols were instituted. As one advocate expressed, “Safety is a challenge for child welfare systems across the country. The buildings may now be ok, but in terms of the dangerous environments they work in (e.g., methamphetamine lab busts, homes with guns, etc.), safety is just an issue nationally.”

Some interviewees said that the litigation also brought about substantial improvements in terms of technological resources, including the provision of cell phones, laptop computers, and personal digital assistants, as well as the development of well functioning, user-friendly automated information systems. As one monitor noted, “The consent decree brought a technological focus to the department.” However, a number of interviewees said that staff in their jurisdictions still lacked cell phones, cars, and working computers.

Leadership

The majority of interviewees said that the litigation brought in new and stronger leadership and helped catalyze needed changes in the workforce culture and organizational environment. However, interviewees noted that, over the “life” of a case, gains were made during times of good leadership, and progress often slowed or stalled during times of bad leadership. They also said that leadership turnover has negatively affected the pace of reform, and that having leaders without practical child welfare or human services experience, as well as limited or inadequate leadership in local area offices, posed challenges to the reform efforts.

Several interviewees said that, in hindsight, there should have been a harder push for new and better leadership in their respective jurisdictions, especially in the context of spearheading the reform. Interviewees also noted that, at times, too much attention was paid by the child welfare agency to developing a new mission or vision for the agency, and not enough effort was put into infusing it into policy and practice. As one monitor said, “It takes time to alter a culture—takes time to create a skills-based and knowledge-based environment. Improved interaction between the workforce and management over time is needed to achieve full buy-in.”

Data and research

The majority of interviewees viewed the litigation as playing a substantial role in improving data collection, analysis, and research activities in their jurisdiction, specifically regarding the indicators and outcomes associated with safety, permanency, and well-being. As one child welfare agency representative explained, “Now we're able to collect data that is actually useful. Data and details about cases are management tools that can be used to learn what is actually going on, leading to better outcomes.” Many interviewees also noted that while the quantity of available information had increased significantly, data reports were not consistently being used to inform the field. As one monitor said, “An organizational focus on data would strengthen issues locally. Giving data more priority and monitoring the use of data, especially at the local level, is something that could be done to improve the workforce.”

A training consultant in one jurisdiction commented,

If you don't look at systemic dynamics, you keep monitoring and looking at numbers all the time, but you can't explain why things are this way. It is critical to understand “What are our workforce dynamics?” in order to answer the question, “Why do we have so many placement moves?” You need to understand the systemic dynamic that drives that performance indicator if you want to make a change.

The majority of interviewees noted that, since the litigation, there had been substantial improvement in agency transparency with stakeholders and quality assurance reporting, as well as the initiation of accreditation activities, ombudsman offices, child fatality review boards, and other stakeholder review boards.

The majority of interviewees also noted that the use of research and evaluation to analyze trends and identify and evaluate best practices also increased following the litigation. However, interviewees noted that comprehensive child welfare research has only recently been a focus in the field and that future efforts by child welfare agencies should support more research studies to identify evidence-based practice models. As one monitor explained:

Research has improved in terms of what to measure and how to measure it, but there is work to be done in understanding what to do with the information. We are much better at assessing data today—the solutions side requires more work. (Public agency evaluation staff) are still unclear regarding what to pay attention to, and what trends mean.

Review of Child Welfare System Data

The review by the Children's Rights and NCYL organizations of available data from the 12 jurisdictions ⁴ pertaining to workforce issues supports the commentary made by interviewees regarding progress since the litigation. Table 3 provides a summary of these findings.

In most cases, the litigation ushered in many important improvements since the initial court order, including increases in funding and a number of other critical developments, such as improved staff recruitment, lower caseloads, better supervisory ratios, more training, active quality assurance mechanisms, new technological supports, and greater data collection. The problem of staff turnover was identified as an area that has not been successfully addressed. In addition, there were only moderate improvements seen in regard to staff incentives (such as salary, benefits, and professional development opportunities), the quality of supervision, working conditions, the use of research to inform policy and practice as well as leadership and agency culture.

LIMITATIONS

The interviews were conducted by staff from Children's Rights and NCYL, which are plaintiffs' counsel organizations involved in many of the lawsuits being studied; also, plaintiffs' counsel, who would presumably have a favorable view of litigation, were among the interviewees. However, it should be noted that, as possible, interviewees' commentary was analyzed quantitatively including and excluding plaintiffs' counsel, and there was a remarkable degree of consistency in the comments and recommendations across the various roles of the interviewees. In addition, no child welfare agency representatives were interviewed in three of the 12 cases because consents to participate were not granted; the number of interviewees differed across cases; the length of interviewees' involvement with the cases varied; and the age of the cases varied.

DISCUSSION AND IMPLICATIONS FOR POLICY AND PRACTICE

Interviewees emphasized that specific activities related to improving the workforce need to be front and center of any policy or practice effort to reform a child welfare system, whether court-ordered or not. Interviewees said that, in order to improve the workforce, it is critical to understand agency culture and to identify the underlying causes of problems in the child welfare system. They noted the need to focus intensively on improving the quality of frontline practice through attention to the background and experience of staff hired, the content of training, the supervisory relationship, quality assurance, and data feedback mechanisms. The focus cannot exclusively be on how many staff are hired, how many hours of training staff are offered, or how many case plans have a current date on them, for example.

TABLE 3 Workforce Issues and Progress in the 12 Jurisdictions

Case	When the lawsuit began and shortly thereafter	As of 2007 (or when the case ended)
Angela R. (AR) ^1	The agency had <50% of the social work staff necessary to meet its own standards of ≤20 cases for foster care workers and ≤12–14 cases for protective services workers. The caseworker turnover rate was almost 50%. 46% of children in foster homes did not receive the required twice monthly caseworker visits during their first month in placement, and 54% of the children did not receive a monthly visit in subsequent months.	Caseloads remain higher than the recommended standard in every area of the state. 19% of authorized caseworker positions are unfilled. 55% of children in foster care receive all required monthly visits.
B. H. (IL) ^2	Average caseloads for child welfare workers were between 50 and 60. Average investigation caseloads for child protective services (CPS) workers were 15–17/month. Supervisory ratios were standardized at 10:1 for CPS workers and 8:1 for child welfare workers, but many supervisors were expected to oversee many more workers, often totaling 150–200 cases a month, and had to supervise in multiple offices. Turnover rate among caseworkers was high. There were not sufficient workers with fluency in the language and culture of clients.	Average caseloads for child welfare workers are an estimated 15 cases/worker. However, caseworkers in 7 of the 10 agency subregions report caseloads in excess of the terms of the 1991 B. H. consent decree. Although investigation caseloads for CPS workers are set at ≤12/month by the consent decree and at ≤9/month by agency Best Practice protocols, some workers still report receiving >12 cases/month. Agency staff was cut by 22% between 2001 and 2005, despite an 11% increase in the number of child abuse investigations. The agency has 25% fewer Spanish-speaking child protective investigators on staff than are needed.
Brian A. (TN) ^3	Caseworkers were routinely responsible for >40 children. When vacancies in caseworker positions were considered, caseloads rose to as high as 80 in at least 1 region. 17% of case-carrying positions were vacant. New workers were trained for only 3 weeks, and workers were unable to take advantage of in-service training due to the demands of high caseloads. Turnover among caseworkers was high. In one office, the turnover rate in 1 year was 100%. Supervisors frequently lacked necessary experience and training, and were also often assigned to supervise more caseworkers than they could effectively manage.	95% of case managers have caseloads at or below the standards set in the Settlement. Staff turnover remains a challenge, with the statewide turnover rate at 19%. Filling all case-carrying positions remains a challenge: 63 positions of 819 were vacant. The agency provides competency-based pre-service training for new caseworkers and new supervisors, as well as regular competency-based in-service trainings. Nearly 90% of supervisors completed a newly developed pre-service training, and 97% of all supervisors were compliant with the supervisory ratios established by the settlement.
David C. (UT) ^4	The caseworker turnover rate was >23%. ^5 There were no established caseload standards and the average number of cases/month was 1,537 with only 78 workers on staff. To meet CWLA standards of 15 cases/worker, the agency needed 129 workers. Pre-service training consisted of reviewing a handbook, a computer presentation on background information on the agency and child development, and a 2-day orientation on agency policies and procedures. Subsequent training consisted of a 2-week certification training to be completed within 1 year. The budget for the agency was approximately $50 million.	The caseworker turnover rate is 20.7%. Average caseloads are 13 cases/worker, with 19% of caseworkers carrying 16–19 cases; 2% carrying 20–23; and 1% carrying ≥24 cases. All caseworkers are required to receive 120 hours of training on Utah's Practice Model prior to receiving a caseload and participate in 40 hours of ongoing training each year through the Child Welfare Institute or other in-service training. The budget for the agency for the 2005 fiscal year was $134,254,800.
G. L. (Jackson County, MO) ^6	The average caseworker had 30 cases, and some workers' caseloads was >50 cases. Inadequate training and insufficient clerical and paraprofessional support made high caseloads even more difficult for caseworkers to handle. Caseworkers missed up to a third of required face-to-face visits with children in their placements. According to the Judge on the case, “excessive” caseloads were “the single greatest hindrance” to the overall reform effort. For example, the Judge drew a clear link between high caseloads and the agency's failure to meet the consent decree's minimum standard of 2 caseworker visits to each child/month.	>80% of caseworkers working exclusively with foster children carried caseloads of ≤15 children, and 95% of supervisory workers oversaw ≤6 front-line caseworkers. The Training Unit ensured that all caseworkers were adequately trained before assuming a caseload. 94% of caseworkers received >100 hours of pre-service training, plus 30 hours/year of continuing training. Along with strengthening upper-level staffing by hiring a new Deputy Program Administrator, the department hired 6 additional full-time clerical workers and 6 additional full-time paraprofessional workers to provide administrative and logistical support to front-line caseworkers. The department created new, specialized staff positions to ensure a consistent focus on key performance areas, adding a full-time Educational Advocate and a full-time Resource Development Coordinator. A Quality Assurance unit provided ongoing review of individual cases as well as systemic trends.
Jeanine B. (Milwaukee, WI) ^7	Caseloads averaged >100 cases/case manager and often were higher. Caseworkers received little or no training before receiving a full caseload. Workers in Milwaukee received significantly lower salaries than those working in neighboring counties. Because worker turnover was high, children were often assigned to “vacant zones,” which had no assigned caseworker.	Caseloads average 10 families/case-carrying manager—one of the lowest average caseloads in the country. Hands-on training for new workers has been expanded, and the agency has partnered with the University of Wisconsin–Madison's Master of Social Work program to offer a flexible, part-time degree program for agency employees. Salaries have been increased, and a stepped salary system has been put in place to reward workers for tenure and education. Case manager turnover rate is still 33%, and case managers' average length of employment is only 1.7 years.
Joseph A. (NM) ^8	Pre-service training was not provided for many new workers, and the training that was provided failed to address the development of worker skills. During one 4-year period in the 1980s, workers were not offered any annual training. Records specifying when employees had been hired and when, if ever, they had attended training, were not maintained. Formal supervisory training was not provided. A significant number of caseworkers were not licensed, and some supervisors lacked both relevant graduate-level education and 3 years of social work experience. Worker caseloads were high and increasing. Caseloads for supervisors were high, with a significant number of supervisors being assigned ≥7 caseworkers to supervise. Supervisors frequently had to carry children's cases >60 days in addition to their regular supervisory duties.	84% of new workers completed the entire pre-service training program within 120 days of being hired. 98% of caseworkers received at least 15 hours of annual training. All supervisors received a minimum of 6 hours of training in supervision within 3 months of assuming their position as well as at least 15 hours of annual training. Interviews with staff indicated dissatisfaction with the pre-service training curriculum and a common belief among social workers that the training did not adequately prepare them for treatment social work. Nearly 100% of caseworkers possessed a current, valid license. Fully 100% of supervisors possessed a current, valid license.
Juan F. (CT) ^9	Caseloads were very high due to a shortage of caseworkers. Turnover rate among social work staff had been steadily increasing—in one office ≥33% of the caseworkers had <1 year of experience, and only ~30% had >2 years of experience. Formal training for workers was completely suspended at one point, and training positions often went unfilled because of a lack of funds. Because of staff shortages, social work supervisors were forced to supervise an excessive number of cases and to provide casework services. No reliable data system existed for identifying individual children in care or tracking basic information on them.	The agency has sustained 100% compliance with caseload standards, established through the lawsuit, for CPS, ongoing, foster care and adoption caseworkers. 90% of families whose cases were reviewed had had only 1 or 2 primary caseworkers during a 12-month period, and in 98% of cases that had to be reassigned, a new primary caseworker was assigned within 5 days. The agency's Training Academy is fully funded and staffed with mandated pre-service and in-service training for all workers. 55% of cases reviewed were being handled by caseworkers who were still receiving “poor” or “negligible” supervision. A high-functioning MIS “LINK” system is utilized to track individualized information on all children in the system and to provide regular aggregate reporting. However, some historic data problems remain for children who have been in care for long periods.
L. J. (Baltimore, MD) ^10	Workers and supervisors lacked the basic training necessary to fulfill their responsibilities. Substantial increases in staff were deemed necessary to reduce the high numbers of cases handled by foster care workers. Cases went uncovered for lengthy periods of time due to high worker turnover and unfilled staff vacancies. Caseworkers did not have access to supports to address special needs of children in care. The data system had major gaps and did not support casework.	All new workers are required to pass a competency training and examination. An L. J. Rally/in-service training was held in September 2005 during which approximately 750 child welfare staff were retrained in 26 areas of knowledge and skill. Caseload ratios required by the consent decree have been achieved in all but one category of staff, but they remain higher than the Child Welfare League of America standards. MD CHESSIE (Children's Electronic Social Services Information Exchange) will be on-line in November 2006, providing caseworkers with comprehensive data on children in the system.
LaShawn A. (DC) ^11	50% of authorized social work positions were vacant. Most social work staff were unlicensed. Caseloads were very high with investigators receiving 30 new investigations/month, and ongoing services workers with caseloads of 49 (intensive services branch) and 59 (continuing services branch). No formal pre-service or in-service training for social work staff existed. Basic working conditions were a serious problem. Concerns included a lack of telephones, vehicles, parking, and staff security.	The agency's vacancy rate is 4.72%. All social work staff are required to be licensed. Workers hired without a license are not kept past the probationary period if they do not become licensed in a short period of time. 80% of investigators, 72% of case-carrying workers (in-home and foster care cases), and 84% of adoption workers have appropriate caseloads. A formal training institute is in place with requirements for pre-service and in-service training. However, not all staff receives the training, despite the fact that it is mandatory. Working conditions have improved. Workers have access to telephones, copiers, cars, and computers with an automated data system for case management.
R. C. (AL) ^12	Social work staff turnover and vacancy rates were very high. Very few workers were BSW- or MSW-level social workers. Caseloads averaged from 40–50 cases/worker, and no caseload standards existed. There was no formal comprehensive training program for social work staff.	Staff turnover rate is 17.5%; the vacancy rate is 2%. 40% of frontline staff are license-eligible social workers. As a result of the lawsuit, caseload standards were established of 1 worker/18 open family services cases, 18 children in foster care, or 22 children available for adoption. 60% of workers have caseloads meeting these standards. Although a formal training program called ACT is in place with requirements for pre-service and in-service training, nearly half of caseworkers in larger counties, such as Montgomery County, are not receiving the basic ACT training.
Ward (Broward County, FL) ^13	Caseloads were high, with some caseworkers carrying caseloads that were 8 times the national standards. Annual turnover rate among caseworkers/supervisors was 80%. Many children had no caseworker or had no caseworker visits for several months. Statewide funding for FY 1998–1999 was $9,823/child.	Although caseloads have decreased to an estimated 21 cases/worker, statutory guidelines require caseloads to be lower (14–17 cases/worker). Yearly turnover rate is 12%, with a vacancy rate of zero. Statewide funding for FY 2004–2005 was $17,966/child.
Note. CWLA, Child Welfare League of America; MIS “LINK,” DCF management information system.

The problem of staff turnover was identified by many interviewees as an area that had not been successfully addressed. Interviewees also noted only moderate improvements in regard to staff incentives (e.g., salary, benefits, professional development opportunities); the quality of supervision; working conditions, including safety issues; the use of research to inform practice; and leadership and agency culture. Interviewees noted that some improvements were difficult to sustain with leadership changes and the changing political and economic climate.

Overall, the results of this study emphasized that workforce reform efforts must be comprehensive and multi-faceted. Focusing solely on training or caseloads alone will not result in workforce improvements that are meaningful and long lasting. Interviewees' commentary, when taken together, provides eight guiding strategies for addressing workforce challenges, whether as a child welfare administrator, policy maker, litigator, or other stakeholder. These eight strategies follow, with some of the most important activities to help support these strategies that were identified by interviewees.

Focus on Leadership, Leadership, Leadership

Consider the following activities:

	Placing the highest priority on identifying and retaining a strong leadership team within the child welfare agency by conducting a national search for agency leaders and managers with child welfare experience and relevant education to ensure the best possible candidates;
	Drawing upon national child welfare expertise to help shape the vision, mission and philosophy of the agency;
	Ensuring that agency leaders actively engage frontline staff by soliciting their feedback and periodically shadowing their work activities to gain a realistic perspective regarding the complex challenges they face and address every day; and
	Engaging local area office leaders and management teams in the reform efforts (not just headquarters), as well as caseworkers and supervisors, to encourage greater buy-in and comprehensive culture change.%

Focus on Retention First, Then Recruitment

Consider the following activities:

	Thoughtfully planning and sequencing the implementation of reforms. For example, an agency cannot add 150 new staff without considering and planning for the various implications of a large-scale hiring, and without ensuring that there is a consistent practice model in place;
	Conducting studies of turnover to better understand the specific factors driving it and inform efforts to improve retention;
	Ensuring that workers have the practical resources necessary to do their jobs, such as desks, computers, cars, and other supports;
	Ensuring that the human resources function is connected to the agency mission and made better aware of agency needs and goals; and
	Providing a realistic job preview that details the work of the organization well to prospective staff.

Pay Attention to Salaries, Benefits and Professional Development Opportunities

Consider the following activities:

	Conducting studies to compare agency salaries to those in other states or counties to make the case to the executive branch and legislature for pay increases and the development of more competitive pay scales;
	Creating specialized, higher-pay positions to attract and retain specially-skilled staff (such as the creation of a sexual abuse unit staffed by staff with advanced degrees and experience in the assessment and treatment of children with sexual abuse histories); and
	Including private community service provider agencies in the problem-solving efforts to understand and address the impact on the system of salary differentials between the public and private sectors.

Improve the Quality of Supervision, Not Just the Supervisor/ Worker Ratio

Consider the following activities:

	Developing supervisory competencies and using them to inform decisions regarding promotion of caseworkers and as benchmarks to monitor supervisor performance;
	Placing supervisors directly in the field with caseworkers to enhance the supervisory relationship and the timeliness of decision-making;
	Requiring supervisors to attend the same trainings prior to caseworkers to ensure effective transmission of classroom-based knowledge to the field;
	Developing comprehensive mentoring programs for supervisors; and
	Creating a managerial level dedicated solely to providing supervision to field supervisors.

Improve the Quality of Training, Not Just the Number of Training Hours

Consider the following activities:

	Ensuring that the training curriculum focuses on meaningful competencies, is based on the most up-to-date information available about best practices, and supports the monitoring of staff performance in relation to training competencies;
	Retraining the entire workforce (not only newly hired staff) when implementing a new policy or practice model;
	Drawing upon the resources of local universities to help develop and implement training;
	Ensuring that social work programs are effectively preparing students for careers in child welfare, and providing stipends for staff who pursue social work degrees; and
	Developing supervisory training models that are compatible with the training provided to caseworkers to ensure greater congruence between caseworker and supervisor philosophy and practice.

Establish Caseload Standards that Reflect a Real Analysis of Workload

Consider the following activities:

	Evaluating workloads so that caseload limits can more appropriately be set;
	Assessing and allocating funding for clerical and paraprofessional support staff;
	Reorganizing staff assignments to allow for job sharing and teaming on cases;
	Locating offices in the neighborhoods where staff are working so that workers are closer to the children and families they serve; and
	Developing specialized staffing departments, including over-hire units designed to satisfy short-term, seasonal, or other workload challenges by maintaining a cadre of trained staff who can fill vacancies as they arise, or through the hiring of part-time or contract staff to fill vacancies caused by planned leaves or hiring freezes.

Provide Safe Working Conditions and Technological Supports for Workers

Consider the following activities:

	Conducting a comprehensive study of working conditions, including safety needs and physical plant issues;
	Soliciting worker feedback when addressing safety or physical plant issues and when developing or refining technological supports to ensure usefulness;
	Allowing for teaming on difficult cases; and
	Decentralizing agency offices to ensure that workers are closer to necessary resources and supports in emergencies.

Collect More and Better Data to Improve Accountability and Expand Research

Consider the following activities:

	Conducting staff focus groups to ensure buy-in to changes in data collection processes and the use of data to influence training, supervision, and policy and practice decisions;
	Improving the availability of data reporting by office, supervisory unit and individual caseworker, and issuing regular regional and unit-level reports to provide information to staff and stakeholders to allow for ongoing comparison and improvement;
	Ensuring that quality assurance units rely on comprehensive data reflecting the quality of services, client satisfaction with services and systemic factors that impact outcomes;
	Utilizing quality assurance reports as tools for supervision and targeted reforms; and
	Collaborating with state and local universities to facilitate comprehensive analysis of agency programs and outcomes.

This study's findings offer a number of recommendations and strategies for improving the child welfare workforce. Although this project examined workforce reforms taking place in the context of class action litigation, many, if not all, of the recommendations are applicable to reform efforts occurring outside of litigation (ACF, 2004; AECF, 2003; IASWR, 2005). These recommendations can be considered whether workforce reforms are being made administratively, legislatively or pursuant to a court order.

CONCLUSION

The recruitment, preparation, support and retention of public and private child welfare staff working with abused and neglected children and families are important and ongoing concerns. Class action litigation focusing on reforming public child welfare systems across the country has often included efforts to improve the child welfare workforce.

Interviewees indicated that progress has been made to improve the child welfare workforce in these jurisdictions and that work remains to be done. The vast majority of all interviewees said that there would not have been improvements without the litigation or that improvements would not have been as significant. Improvements noted by interviewees since the litigation included increases in funding for the system and improvements in staff recruitment, caseloads, supervisory ratios, training, quality assurance mechanisms, technological supports, and data collection. Many interviewees said that these workforce improvements had begun to translate into improved case practice and better outcomes for children and families. However, some interviewees said that workforce improvements that were made as a result of litigation had yet to result in improved outcomes. Much work remains to be done to improve staff retention, staff incentives, the quality of supervision, leadership and agency culture, and the use of data and research.

This study makes clear that additional supports for the workforce are needed in order to improve outcomes for children and families involved with child welfare systems across the nation. Workforce reform efforts should be comprehensive and multi-faceted, and not focus solely on increasing the number of training hours or reducing caseloads, as these types of one-dimensional actions are not likely to result in systemic improvements that are meaningful and sustainable. Interviewees' comments and recommendations for moving forward provide a useful framework of the eight components necessary for any reform effort to strengthen the child welfare workforce, and, indeed, the system as a whole.

Notes

This study was conducted by the authors when they were employed at Children 's Rights. The full report is available at www.childrensrights.org/wp-content/uploads/2008/06/improving_the_child_welfare_workforce_full_report_feb_2007.pdf. The authors wish to ac- knowledge the contribution of Madelyn Freundlich, under whose supervision this study began. The authors also acknowledge the valuable contribution to this work made by the National Center for Youth Law, 405 14th Street, 15th Floor, Oakland, CA 94612. Available at: www.youthlaw.org/

Note.¹The term workload refers to the amount of time it takes to complete all case-related tasks (e.g., transporting children to family visits, developing a case plan, meeting with foster parents). DHS = Department of Human Services; DCFS = Department of Children and Family Services (Arkansas)/Division of Child and Family Services (Utah).

¹ Angela R. Complaint (1991); Kelly, 2005; Field Staff Summary (2006); Statewide Compliance Outcome Report (2006). It should be noted that the Angela R. case ended in 2001; data indicate that progress has not been sustained in certain areas.

²AFSCME Council 31 (2006); Final Consolidated Report of Rule 706 Panel of Experts (1990); Illinois Department of Children and Family Services (2004); Report for Court on Child Protective Services (1990).

³ Brian A. Complaint (2000); Monitoring Report of the Technical Assistance Committee (TAC) (2006).

⁴ David C. Complaint (1993); Child Welfare Policy and Practice Group (2005); Utah Children (1988); Utah State Department of Social Services (1985); DCFS Database (2005); DCFS (2004); Utah Legislature (1993); Utah Child and Family Services (2005).

⁵Turnover data was not available in 1993; the earliest available data on turnover rates is from 1998.

⁶ G. L. Contempt Motion (1985); G. L. Complaint (1977); G. L. Contempt Order (1992); Children's Division Quality Assurance Unit (2005); Monitoring Committee (2005). G. L. was filed in 1977, and contempt motions were filed in 1985 and 1992. The Court's 1992 Opinion and Order finding Defendants in contempt of court for failure to comply with the terms of the consent decree paid particular attention to workforce issues.

⁷ Jeanine B. Complaint (1993); Department of Health and Family Services (2006); Bureau of Milwaukee Child Welfare (BMCW) (2006).

⁸ Joseph and Josephine A. (June 1999; December 1999); Plaintiffs' Post-Trial Memorandum (1988).

⁹ Juan F. Complaint (1989); Juan F. (2005; 2006).

¹⁰ L. J. v. Massinga (1988); L. J. Complaint (1984); Department of Human Resources (2006).

¹¹Center for the Study of Social Policy (CSSP) (1990; 1991; 2005).

¹²Alabama Department of Human Resources (DHS) (November 2004; January 1994; April 2006; May 2006).

¹³ Ward Complaint (October 1998); Office of Program Policy Analysis & Government Accountability (2006).

1. For the sake of brevity the term court orders is used throughout this article as a general term to encompass all the court-ordered remedies represented in this study, although in many instances the remedies may have been by consent. Despite the nomenclature, the common thread is that all are/were court-enforceable, court-ordered remedies.

2. Complaints are filed by a person or entity claiming legal rights against another; they are pleadings and must properly state the factual as well as legal basis for the claim, name both the party making the claim and all defendants, and state what damages or performance is demanded.

3. Court orders in class action child welfare litigation typically require that periodic “monitoring reports” be prepared by third-party child welfare experts who are chosen by the parties to assess an agency's compliance with the terms of the agreement.

4. Reviewed documents included initial complaints, early and recent monitoring reports, and other documentation of the status of the child welfare system in each jurisdiction.