The Florida Study of Professionals for Safe Families—Final Report 2015-2020

The Florida Study of Professionals for Safe Families Final Report 2015 – 2020 September 2020

Funded by

Contents Introduction .......................................................................................................................................................... 4 Background ....................................................................................................................................................... 4 Recruitment .................................................................................................................................................. 4 Retention ...................................................................................................................................................... 5 Communication ............................................................................................................................................ 6 Summary of Findings ............................................................................................................................................ 7 Turnover ........................................................................................................................................................... 7 Data: Agency Turnover - Full Cohort ............................................................................................................ 7 Data: Agency Turnover - By Role ................................................................................................................ 12 Training and Transition ................................................................................................................................... 12 Implications ................................................................................................................................................ 16 Work Experiences ........................................................................................................................................... 16 Caseload Responsibilities ........................................................................................................................... 16 Implications ................................................................................................................................................ 19 Client Violence ............................................................................................................................................ 20 Organizational Culture................................................................................................................................ 22 Data: Supervisor and Co-Worker Social Support ........................................................................................ 23 Worker Responses .......................................................................................................................................... 24 Implications ................................................................................................................................................ 27 Job Satisfaction ............................................................................................................................................... 27 Scale Validations ............................................................................................................................................. 27 Child Welfare Provider Stigma Inventory ................................................................................................... 27 Supervision Practice in Human Service Scale ............................................................................................. 28 Intimate Partner Violence Responder Collaboration Survey ..................................................................... 28 Recommendations .............................................................................................................................................. 29 Pre-service Training .................................................................................................................................... 29 Supervisor Competencies ........................................................................................................................... 29 Agency On-boarding ................................................................................................................................... 30 Culture of Self-Care .................................................................................................................................... 30 Future Directions ............................................................................................................................................ 30 Future Research.......................................................................................................................................... 30 Legislative Priorities (SB 1326) ................................................................................................................... 31

2|Page

Florida Institute for Child Welfare

Appendices Appendix A - Agency Reports…………………………………………………………………………………………………….…………..33 Appendix B - Research Briefs………………………………………………………………………………………………………….……..84 Appendix C – Scales……………………………………………………………………………………………..…………….….……………112 Appendix D – Bibliography……………………………………………………………………………………………………….….……..119

Table of Figures Figure 1: Participation ................................................................................................................................... 5 Figure 2: Response rates ............................................................................................................................... 6 Figure 3: Agency departure........................................................................................................................... 7 Figure 4: Departure reasons ......................................................................................................................... 8 Figure 6: Departure details – supervision ..................................................................................................... 9 Figure 5: Departure details – job responsibilities ......................................................................................... 9 Figure 8: Departure details – other career opportunities .......................................................................... 10 Figure 7: Departure details – agency environment .................................................................................... 10 Figure 9: Departure details – personal reasons .......................................................................................... 11 Figure 10: Departure by role ....................................................................................................................... 12 Figure 11: Departure reasons by role ......................................................................................................... 12 Figure 12: Perception of caseload size........................................................................................................ 18 Figure 13: Client violence by category ........................................................................................................ 21 Figure 14: Social support............................................................................................................................. 23 Figure 15: Responses to work responsibilities ............................................................................................ 26 Figure 16: Job satisfaction........................................................................................................................... 27

3|Page

Florida Institute for Child Welfare

Introduction

The Florida Study of Professionals for Safe Families (FSPSF) is a prospective longitudinal study of Florida’s child welfare workforce. The primary focus of the study examines turnover and retention decisions among frontline workers. The FSPSF project was funded by the Florida Institute of Child Welfare and was led by Dr. Dina Wilke (Principal Investigator), Dr. Melissa Radey (co-Principal Investigator), and Dr. Lisa Magruder (Project Manager/Post-Doctoral Scholar). Other contributors included Dr. Philip Osteen, Dr. Erin King, Dr. Stephanie Kennedy, Sarah Rakes, Carmella Miller, Caitlin Nolan, Cassandra Olson, and Lauren Stanley.

BACKGROUND

In 2014, the Florida Legislature passed SB1666, signed by Governor Rick Scott, was an overhaul of child welfare statutes. Among the many components, this bill created the Florida Institute for Child Welfare (FICW), which included several expectations to assess the health of the child welfare workforce.

Recruitment and retention for child welfare professionals are widespread issues for the Department of Children and Families (DCF) and the community-based care lead organizations (CBCs). High staff turnover puts vulnerable children at greater risk for recurrence of maltreatment, impedes timely intervention referrals and, ultimately, delays permanency. 1 The Florida Study of Professionals for Safe Families (FSPSF) surveyed a cohort of newly hired dependency case managers and child protective investigators every six months for approximately 3.5 years. The study focused on individual, organizational, and community influences on child welfare employee retention. This statewide study examined worker personal characteristics (e.g., educational background, family history, self-esteem, etc.) worker experiences and responses (e.g., stress and burnout, work/family balance, social support, and coping, etc.), and organizational influences (e.g., organizational climate, supervisory practices, and caseload responsibilities etc.).

The FSPSF utilized three broad strategies to answer several different research questions. First, respondents were surveyed every 6-7 months with a core instrument. Second, in addition to the core instrument, three in-depth modules were rotated during the data collection period. Modules included a focus on 1) supervision and organizational functioning (assessed at waves 2, 5, and 8); 2) work/family life balance (assessed at waves 3, 6 and 8); and 3) mental health (assessed at waves 4 and 7). The intent of this strategy was to gain a deeper understanding of key areas of personal or organizational characteristics that may impact job satisfaction and retention. Finally, qualitative interviews were used to explore emerging topics in more depth and included interviews on the transition experiences from pre-service training into caseload responsibilities and experiences of client-perpetrated violence.

Recruitment

FSPSF project staff recruited all Child Protective Investigators (CPIs) and Dependency Case Managers (CMs) who were in pre-service training between Sept. 1, 2015 and December 31, 2016. We followed this cohort of new hires for 45 months, or about 3.5 years following training. Participants were followed even if they left their child welfare positions during the study timeframe. This strategy was critical to understanding employment outcomes for those who left their initial CPI/CM positions. Participants were recruited during their pre-service training, a mandatory training for all new hires not currently holding a Florida certification in the job for which they have been hired. Overall, 100 percent of child welfare administrative units across the state of Florida agreed to be part of the study. This included the 17 community-based care lead organizations, 6 regions for the Department of Children and Families, and 6 sheriff offices.

4|Page

Florida Institute for Child Welfare

Face-to-face recruitment was conducted during visits to pre-service training classes in an effort to build credibility and connection with those considering participation. These visits were used to explain the project to trainees and seek their consent for participation. Trainees also provided personal phone and email contact information so that study engagement was completed without using employer resources.

When a trainee provided their consent to participate, they were sent the baseline survey. All workers who provided consent and completed a baseline survey became part of the FSPSF cohort.

Data: Recruitment

Between September 2015 and December 2016, FSPSF staff visited 160 pre-service training classes to recruit participants. Because of on-going communication and rapport built with training staff, we are confident we went to 100 percent of scheduled pre-service training classes. All told, there were 1,803 trainees in the sessions we visited, and 1,662 trainees provided consent and agreed to participate (92.1%). Of those, 1,500 (90.3%) returned baseline data representing 83.2% of all eligible workers across Florida. There were regional differences, however. Combining the participation rate for CPIs and CMs, the Central region had the highest Figure 1: Participation rate of participation Participation by Region (88.2%), followed by 100.0% the Northeast Region 90.0% (86.0%), the 80.0% Northwest Region 70.0% 60.0% (80.2%) and the 50.0% Suncoast Region 40.0% (83.9%). The 30.0% 20.0% Southeast and 10.0% Southern regions had 0.0% the lowest Northwest Northeast Central SunCoast Southeast Southern participation rates % Consent % Complete (74.4% and 70.1% respectively) As a result, the experiences of workers in these two regions may be underrepresented in our overall findings. We did not collect information from eligible workers who declined to participate and are unable to assess if they were systematically different from those who did. Figure 1 shows the distribution of eligible workers who provided consent and baseline completion by region.

Retention

The FSPSF cohort was divided into quarterly panels based on study recruitment date in order to ensure that participants were surveyed at the same general point in their employment tenure. The FSPSF had a robust distribution and reminder protocol to encourage survey completion. Initial contact was made with participants via text message approximately one week prior to survey distribution. This was to inform them of the upcoming distribution and to give them an opportunity to get in touch with project staff if they had any changes to their contact information. Surveys were distributed via email on Saturday mornings to not interfere with work obligations. Following distribution, a second text message was sent along with an email from the PI to alert participants in the event the survey was sent to a spam folder or not distributed by the email provider. Surveys remained available for completion for 50 days.

5|Page

Florida Institute for Child Welfare

Text message reminders were sent on days 20 and 27, and email reminders were sent on days 7, 14, 21, and 28.

Table 1: Incentive by wave of data collection

Wave 2

Wave 3

Wave 4

Wave 5

Wave 6

Wave 7

Wave 8

Survey Interval

6 months

6 months

7 months

6 months

7 months

6 months

7 months

Time since pre-service

6 months

12 months

19 months

25 months

32 months

38 months

45 months

$25

$40

$40

$50

$50

$60

$75

Incentive amount

In recognition of the time required for their involvement (each survey took approximately one hour to complete), a monetary incentive, in the form of an e-gift certificate, was provided to all participants who completed a survey. In accordance with best practices in longitudinal research 2 incentives were gradually increased to maintain participant interest and commitment. Table 1 includes a description of the amount of incentive by wave.

Data: Response Rates

Response rates for each wave were very high. Across all waves of data collection, response rates averaged 80.6 percent and ranged from 77 percent to 87 percent (Figure 2). Further, 62 percent (n = 929) of participants responded at every wave of data collection.

Communication

Figure 2: Response rates

Response Rate by Wave of Data Collection 100.0 80.0 60.0 40.0 20.0

0.0 Several strategies were undertaken to W2 W3 W4 W5 W6 W7 W8 inform stakeholders about key findings from the FSPSF. First, following each wave of data collection, summary reports were provided to executive leadership of the CBCs, sheriffs’ offices, and DCF regions (Appendix A). In these reports, key findings on turnover, reasons for departure, and workers’ perceptions of their employment were presented at three levels of analysis. Findings were provided for all child welfare workers across Florida who responded at a given wave of data collection.

Findings were also presented by type of employee (e.g., DCF PI, case manager, and sheriffs’ office PI) and by specific agency. For example, a given CBC would receive a report with columns of data for all child welfare workers, all case managers, and their specific administrative unit. This was done to allow for benchmarking the responses from an individual unit with similar workers across Florida and with all child welfare workers regardless of role. DCF regional leadership received reports by circuit. The agency reports summarized results from a wave of data collection and were provided semi-annually.

Second, research briefs have been prepared quarterly and distributed through the FSPSF mailing list and the Florida Institute for Child Welfare website. Each report is a brief summary of current analyses and is written for practitioners. Topics included training and transition experiences of new workers, perceived

6|Page

Florida Institute for Child Welfare

changes in mental and physical health for workers over time, self-care practices of workers and their impacts on employment experiences, and reasons for departure over time, among others. All research briefs are available in Appendix B.

Summary of Findings

Below is a summary of findings highlighting key areas of focus for the FSPSF including findings on turnover, experiences with training and transition to casework, caseload responsibilities, client violence, and worker well-being.

TURNOVER

At each wave of data collection, participants were asked if they remained employed in their original child welfare agency. Those who said “no� were asked a series of follow up questions including their primary reason for departure and new employment information as applicable. The following results represent findings from the workers who indicated leaving their original hiring agency. However, this definition likely does not capture the true impact of departure on child welfare caseloads for two reasons. First, many organizations providing child welfare services also provide other social or law enforcement services. Therefore, some workers could leave their child welfare positions, yet indicate they still worked for their original hiring agency. For example, a CPI at the Department of Children and Families could have moved to a position in Economic Self-Sufficiency and indicated they still worked for DCF. Second, this definition does not include workers who left frontline casework for a promotion or related opportunity in their agency. Accordingly, the percentage of agency departure at any given wave may slightly underestimate caseworker turnover for clients.

Data: Agency Turnover - Full Cohort

By wave 8 - approximately 3.5 years post-baseline – 81 percent of workers had left their original child welfare agency. Figure 3 shows the percentage of employees reporting agency departure by wave of data collection and includes a marker for cumulative Figure 3: Agency departure departure. Overall, the first 18 months represent the Agency Departure by Wave most common time for departure with nearly 57 100.0% 90.0% percent of new workers 80.0% reporting agency departure 70.0% during that time period. The 60.0% single greatest window for 50.0% 40.0% departure was between 730.0% 12 months of employment, 20.0% when 23 percent of the 10.0% cohort indicated agency 0.0% W2 W3 W4 W5 W6 W7 W8 departure. Importantly, about the same proportion % Turnover Cumulative of workers left during the first year of the study (41.0%) as left in the remaining 2.5 years (40.3%). The first year includes only 9-10 months of casework as the first 2-3 months are reserved for classroom-based training.

7|Page

Florida Institute for Child Welfare

At each wave of data collection, when a worker indicated job departure, they were asked to provide their primary reason for leaving. Overall, there were 39 options workers could select from and those were collapsed into five categories. The first category included reasons associated with workers’ perceptions of their job responsibilities, such as “too many hours” or “caseload was too complex.”

A second category included reasons associated with workers’ perceptions of the agency environment such as having “unreasonable expectations” or “dissatisfaction with salary and benefits.” Workers’ perceptions of supervision comprised the third category of departure reasons including having a “bad work ethic” or “not available.” Examples of personal reasons included “birth or adoption of a child” or “family relocation”, and other career opportunities included options such as “promotion at a different agency” and “better job opportunity.” During the first four waves of data collection, workers were given an “other” option Figure 4: Departure reasons although this was eliminated Departure Reason by Wave beginning at 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% wave 5. Figure 4 W2 shows the primary W3 reason for leaving W4 among departed W5 workers over time.

W6 Overall, workers W7 primarily left for W8 reasons associated with their job Job Responsibilities Agency Environment Supervision responsibilities or Personal Reasons Career Opportunity Involuntary Departure the agency Other environment.

The reasons for departure and associated detail for each category were aggregated across all waves of data collection through wave 5, which is approximately two years post baseline. After wave 5, the number of people departing was too small to meaningfully examine detailed reasons for departure.

Among all workers who left within the first two years of their employment, the most frequently cited reasons were associated with their job responsibilities (28%) followed by agency environment (24%), personal reasons (15%), other career opportunities (13%), supervision (10%), and involuntary departure (5%). Other reasons were identified by 6 percent of those who left their agency. Figures 5-9 present the detailed reasons for departure within each category. The question is set up such that workers can only provide one primary reason for departure. Each chart represents a category of departure and presents the percentage identifying a particular reason within that category.

8|Page

Florida Institute for Child Welfare

Figure 6: Departure details – job responsibilities

Departure Detail: Job Responsibilities n = 208

0%

5%

10%

15%

20%

25%

30%

Too many hours Too many cases Too emotionally difficult Caseload too complex Too much travel Too much paperwork Didn't utilize my skills Safety concerns Job not in my field

Figure 5: Departure details – supervision

Departure Detail: Supervision n=72

Had unreasonable expectations Not available Played team members against others Bad work ethic No constructive feedback provided Unable to work with cultural differences Bad-mouthed agency/clients Not culturally sensitive Didn't assign cases fairly Not supportive 0%

9|Page

5%

10%

15%

20%

25%

30%

35%

Florida Institute for Child Welfare

Figure 8: Departure details – agency environment

Departure Detail: Agency Environment n=176

Had unreasonable expectations Showed little concern for workers Too concerned with policies Pushed to act against my better judgment Wasn't provided with needed resources Dissatisfied with salary/benefits Few opportunities for advancement No acknowledgement of good work Insensitive to cultural diversity 0%

5%

10%

15%

20%

25%

30%

35%

40%

Figure 7: Departure details – other career opportunities

Departure Detail: Career Opportunities n = 95

Better work opportunity

Return to school

Promotion at a different agency

0%

10 | P a g e

10%

20%

30%

40%

50%

60%

70%

Florida Institute for Child Welfare

Figure 9: Departure details – personal reasons

Departure Detail: Personal Reasons n = 109

Family relocation

No work/family balance

Medical condition

Birth/adoption of a child

0%

5%

10%

15%

20%

25%

30%

35%

40%

45%

Finally, the FSPSF followed the full cohort of workers for the 3.5 years of data collection, including those who departed their original hiring agency. This enabled us to identify the number of workers who left one child welfare agency for another child welfare position. Workers were asked to identify the title of their new role (e.g., adoptions specialist, trainer, etc.) to verify their continuing employment in the child welfare system. Overall, while 81 percent of the participants left their initial hiring agency, about 41 percent of the cohort remained working in a child welfare role 3.5 years post baseline.

11 | P a g e

Florida Institute for Child Welfare

Data: Agency Turnover - By Role

Agency departure and the reasons identified for departure were also examined by worker role, as worker responsibilities vary across the child welfare system. Roles included CPIs employed by DCF, CPIs employed by a county Figure 10: Departure by role sheriff’s office, and dependency CMs Departure by Wave and Role employed by local 30.0% agencies contracted with 25.0% a CBC. Figure 10 provides 20.0% departure data by wave of 15.0% data collection. Overall, 10.0% more CMs left their 5.0% agency (89%) than did 0.0% CPIs from DCF (77%) and W2 W3 W4 W5 W6 W7 W8 CPIs at sheriffs’ offices DCF PIs Sheriff's Office PIs Case Managers (69%). Although there were few statistical differences between roles over time, departure spiked more dramatically during the 7-12-month time period (wave 3) for the DCF CPIs as 28 percent reported leaving the agency during that time.

Figure 11: Departure reasons by role Figure 11 presents the reasons for departure by Departure Reasons by Role role aggregated through wave 5. Most reasons for Case Manager departure were similar Sheriff PI across roles. For example, a similar proportion of DCF PI workers in each role 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% identified job Job Responsibilities Agency Environment Supervision responsibilities as the most Personal Reasons Career Opportunities Involuntary Departure common reason for Other leaving. Supervision and personal reasons were also identified by similar proportions across role. However, while the agency environment was the second most identified reason for departure, it was reported by more DCF CPIs compared to case managers.

Further, more sheriff’s office CPIs and case managers left for other job opportunities compared to DCF CPIs, and although involuntary departure was a relatively uncommon reason for departure, more CMs were terminated than those in other roles.

TRAINING AND TRANSITION

The previous findings on agency departure identify the critical importance of the first-year window of employment. Newly hired workers begin with an 8-12-week pre-service training program. In Florida, multiple groups of workers typically come together in a centralized location for the classroom-based portion of pre-service training. As an example, training for child protective investigators is generally organized around a judicial circuit, which may bring together workers assigned to as many as five or six 12 | P a g e

Florida Institute for Child Welfare

different county service centers. Dependency case managers have a similar structure of training in a centralized location and providing casework services in different local agencies. This semi-centralized training model may introduce challenges when workers begin to apply classroom learning within individual agency environments. Both quantitative and qualitative analyses on workers’ experiences with their pre-service training and the transition into child welfare casework were conducted.

First, a sub-sample of FSPSF respondents (n = 38) completed telephone interviews addressing their initial employment experiences after approximately six-months post-baseline. Workers expressed a variety of reactions to their training with almost one-third describing it in universally positive terms (n = 11, 29%), one-third describing it in universally negative terms (n = 13, 34%), and the largest number identifying both negative and positive components (n = 14, 39%). CPIs were more likely to rate their experiences as uniformly positive than case managers (43% versus 12%) largely because case managers expressed that the training content was too focused on CPI tasks. From their experiences, workers identified five attributes of pre-service training that were meaningful to their transition to independent casework: 1) structured, non-redundant content relevant to their positions; 2) interactive content delivery; 3) practice with job responsibilities; 4) practice with system protocols; and 5) recognition of a never-ending learning process. The sections below describe each theme and suggest ways to promote meaningful pre-service training experiences.

1.

Structured Relevant Training Content

2.

Interactive Content Delivery

3.

Practice with Job Responsibilities

Workers identified that training content was central to their success, and they desired structured, new, non-redundant, and relevant information. Workers were evenly split on whether they felt the training provided relevant content. A structured schedule of training helped workers see the “big picture” and connect training content, exam material, and field responsibilities. Some workers with child welfare education backgrounds or prior professional experience felt the content was a “refresher” and desired additional field experiences to expand their current knowledge. Workers commonly expressed that training content focused on “theoretical” foundations at the expense of procedural knowledge or focused too heavily on job tasks that did not align with their positions.

Almost universally, workers wanted an engaging training atmosphere that gave them a “feel” for the job rather than the overuse of presentation slides and trainer “scripts.” Approximately one-half of workers described interactive training experiences with fewer workers describing primarily lecture-based experiences. Workers who reported lecture-based experiences requested more interactive training segments to “break up the monotony” and were disappointed when they did not have the opportunity to apply material in “what-to-do situations.”

Workers wanted to practice job responsibilities while in training and desired classroom practice opportunities with case scenarios that exposed them to “the full, the realness, the rawness” of child welfare work. Workers also wanted more field experience. Successful field experiences typically contained three characteristics: finding a certified worker to shadow, completing job tasks in the field (as opposed to clients not being available), and discussing the completed job tasks with the certified worker. Multiple workers mentioned the need for them to be “aggressive” or “hustle” to find a worker willing to let them shadow.

13 | P a g e

Florida Institute for Child Welfare

4.

Practice with System Protocols

5.

Never-Ending Learning Process

Workers desired and appreciated training practice with case documentation, specifically with the necessary computer software systems. Most workers who discussed system protocols identified practice with protocols as a central missing training component. Workers recognized that proper case documentation was a required element of the “nitty gritty” of child welfare work. They desired more guidance with navigating the computer software documentation system and what makes a “good note versus just satisfactory.”

Regardless of the quality of their training experiences, workers generally recognized they could never be fully prepared for child welfare work when they transitioned to independent caseloads. Several workers identified the unpredictability of their work and diverse needs of their caseloads as a primary reason for the need for continuous learning in their jobs. Workers mentioned the importance of their supervisors, mentors, and co-workers as supports for continuous learning. Lack of support for workers made the transition from classroom to casework difficult. These workers felt vulnerable and reported only receiving instructions after making mistakes. Those without support often felt “alone” and discouraged.

A second analysis focused more specifically on agency-based field days during their pre-service training.

Field days are built into the pre-service training curriculum to allow workers time in their agencies to observe or practice skills being presented during training.

Workers’ experiences during field days were uneven. All workers expressed that field days played an important role in their job preparation. However, while almost 50 percent (n = 17) had positive, meaningful experiences, 40 percent (n = 15) did not. Those with positive experiences generally felt that their field days exposed them to critical, realistic job content while those with negative experiences felt frustrated that they received incomplete training and wasted time in agency offices without guidance.

The remaining participants felt mixed about their experiences (n = 6) such that although some field days were helpful, they desired more exposure to job tasks and procedures. Data analysis identified four themes related to the conditions facilitating meaningful field day experiences and preparation for their positions.

1.

Integration of Knowledge and Skills

2.

Structure of Field Days

Participants consistently viewed field days as an excellent opportunity to integrate pre-service classroom learning into the practice setting. Observing mentors and applying content from the classroom-based training enabled newly hired workers to improve the skills needed for their positions.

Likewise, participants emphasized that field experiences provided exposure to the reality of the child welfare context, which facilitated their development of realistic job expectations. Regardless of the quality of the experience, participants expressed that field days could offer a supervised forum in which they could “see what you learned and actually put it into play.”

Newly hired workers expressed that having structure within field days contributed to the quality of learning experiences. Most participants with meaningful experiences had field days with a defined purpose, assigned tasks, and an experienced mentor assigned to them. Tasks included shadowing the mentor while conducting a home visit, attending court, or completing administrative tasks necessary for documentation. Alternatively, workers whose field days lacked structure expressed frustration and regret that they “just sat” during their time at the agency due to unassigned mentors and fluctuating

14 | P a g e

Florida Institute for Child Welfare

schedules. In these circumstances, agencies were unprepared to provide appropriate experiences for new workers.

3.

Functioning in Disrupted Environments

4.

Connection to Mentors

The work setting shaped participants’ field days, and at times disrupted meaningful experiences. For some workers, high caseloads, the fast-paced nature of the work, the priority of timely investigations, and timelines for case closures translated to mentors not being available. Typically, workers did not blame mentors who were unable to attend to them due to their own job responsibilities. In fact, despite frustration in having limited opportunities to observe or participate in critical job tasks, study participants valued gaining a realistic view of job demands.

The quality and level of engagement between the newly hired worker and mentor played a key role in the quality of the field training experiences. Several study participants with meaningful experiences noted the patience of their mentors; they connected with their mentor. Alternatively, workers without meaningful experiences described lacking approachable mentors. In addition to the structure and content of field days, the quality of interaction between the worker and the mentor contributed to participants’ ability to ask the questions necessary to learn job tasks and feel welcomed in the field.

A final quantitative study on the transitions experiences of newly-hired workers focused on availability of specialized mentoring, the number of cases received in the first week providing services, and perceptions of the consistency between agency practice and pre-service training as predictors for departure (turnover) within the first six months of employment.

Predictors of Early Turnover

Overall, 18 percent of study participants left their agencies within the first six months (n = 235) following pre-service training. Early leavers were older and had more prior work experience in any field than those who remained in their child welfare jobs. There were no significant differences in coping strategies, satisfaction with pay or benefits, support received from supervisors and co-workers, or previous child welfare work experience between the two groups. However, early leavers and those who remained significantly differed on all three measures of transition experiences. Fewer early leavers received specialized mentoring when beginning their caseload responsibilities than those who remained (56% and 70% respectively). Caseload sizes for all workers in their first week of casework ranged from 0 cases assigned to 27 cases, and on average, early leavers reported a higher initial caseload than those who remained (3.2 cases vs. 2.5 cases respectively). Early leavers also reported more discrepancies between information provided in training and actual agency practice; 54 percent of early leavers indicated that agency practice was rarely or not at all consistent with training information compared to 30 percent of those who remained.

Workers’ transition experiences also predicted the likelihood of early departure. After controlling for a worker’s personal background, each additional case assigned the first week of casework increased the likelihood of departure by the six-month period by 8 percent. This suggests that a worker assigned 10 cases in the first week following training was 80 percent more likely to leave than a worker assigned no cases. Further, compared to workers who indicated training was completely consistent with agency practice, those who said it was rarely consistent were 90 percent more likely to leave, while those who said agency practice was not at all consistent with training were four times more likely to leave within the first six months of employment.

15 | P a g e

Florida Institute for Child Welfare

Implications

As previously reported, there was as much agency turnover in the first-year post baseline as in the subsequent 2.5 years of follow up. The transition period from a centralized classroom training experience to agency-based casework is often a stressful one. Policies and practices that support workers during this transition time are an important strategy to reduce early turnover. Specific recommendations include: •

Incorporate opportunities for agency-specific policy and procedure training during the standardized pre-service training calendar.

•

Identify agency-based liaisons to serve as a single point of contact for training personnel in order to coordinate the field day experience of trainees. This could include planning daily field day activities, assigning trainees to effective mentors, coaching mentors on the needs of trainees, and providing feedback to training personnel on the continued learning needs of trainees based on field day performance. Liaisons could also inform curriculum development and delivery and ensure relevance to agency-based practice.

•

Identify specific agency representatives to serve as educators, mentors, or advocates for new hires transitioning to from pre-service training to casework.

•

Carefully monitor the early workload of new hires and develop processes to ensure agency guidelines regarding case assignment are being followed.

•

Structure classroom and field training to specific positions (i.e., CPI, CM), focused on real-world job tasks and case documentation, and with less emphasis on theoretical foundations of child welfare.

•

Support training processes that 1) incorporate a variety of learning modalities; 2) limit lecturebased, scripted delivery of content; and 3) facilitate interactive, discussion-driven training sessions.

•

Identify approachable and supportive agency professionals to serve as mentors and provide training on best practices for teaching and coaching new employees. Consider agency policies that allow experienced workers to mentor as part of their job responsibilities in exchange for reduced caseloads or other benefits.

•

Develop checklists for mentors and trainees to use during field training days and transitioning to casework to optimize learning experiences.

•

Have alternative activities prepared for trainees whose training day activities become disrupted due to unforeseen circumstances (e.g., alternative mentors, training about agency-based policies and practices).

WORK EXPERIENCES Caseload Responsibilities

The Child Welfare League of America (CWLA) recommends different caseload sizes based on the family’s needs. 3 For example, for initial maltreatment investigations, the CWLA recommends no more than 12 active cases per month, while for workers providing ongoing family support, CWLA recommends no more than 17 active families. Smaller caseloads are recommended for families with higher intensity needs or services. In the FSPSF, at each wave, workers were asked for information regarding their caseload responsibilities. This includes questions about the size and difficulty of their caseloads, the number of visits made, the number of hours spent driving for work responsibilities in the previous seven days, and the number of days in the past month that workers began work early, stayed late, or worked 16 | P a g e

Florida Institute for Child Welfare

on a day off. Caseload size was created by summing the number of cases workers indicated they had in the following categories: 1) regularly assigned cases; 2) courtesy cases from other Florida counties; 3) courtesy cases from other states; 4) current cases placed in another state; and 5) other types of cases.

All data presented below comes from workers who indicated that they remained working in child welfare at a given wave and that they carried a caseload (Table 2). a

Table 2: Eligible child welfare workers by wave

Wave

W2

W3

W4

W5

W6

W7

W8

Eligible Workers

1,068

836

660

538

456

396

333

Data: Caseload Size and Difficulty

Over the 3.5 years of the FSPSF, workers’ caseload size and perceptions of difficulty were quite stable.

At six months, workers indicated an average caseload size of 14.6 cases. By 12-months, that increased to 16.8 cases per worker but stayed steady for the remaining waves of data collection. The average number of cases across all waves was 16.1 and ranged from 14.6 to 16.8. Workers were asked to identify the number of children on their caseload. This was a bit more variable and ranged from a low of 26.4 children to a high of 31.4, with an average of 29.1 children over all waves of data collection.

Overall, while the low point for both caseload (14.6 cases) and number of children on a caseload (26.4 children) was at the 6-month mark, they were nonetheless high for workers who were within 3-4 months of completion of pre-service training.

Workers were also asked to give their perceptions of the size of their caseload with answer options of “too low,” “too high,” and “about right” (Figure 12). In contrast with expectations, at every wave of data collection, the proportion of workers who indicated their caseload was “about right” was the same or higher than those believing their caseload was “too high.” Further, the difference between those indicating their caseload was too high and those saying it was about right grew over time.

As the study progressed and workers’ responsibilities changed, the number of cases identified by workers varied significantly. In order to best capture the experiences of frontline case workers, we truncated the number of cases reported by workers to 36, or three times the Child Welfare League of America’s recommended caseload of 12 cases per worker. At each wave, beginning with wave 3, between 10-15 individual responses were truncated. Similarly, the number of children on a caseload also included a few extreme outliers. For example, one respondent indicated they were responsible for all children in a certain category and estimated the number of children on their caseload to be well over 1,000. In order to reduce the impact of extreme outliers, the number of children on a caseload was truncated at 500. Across all waves, there were fewer than 10 answers truncated to 500 children.

a

17 | P a g e

Florida Institute for Child Welfare

Figure 12: Perception of caseload size

Perception of Caseload Size by Wave 100.0 90.0 80.0

Percent

70.0 60.0 50.0 40.0 30.0 20.0 10.0 0.0

W2

W3

W4

Caseload: Too High

W5

W6

W7

W8

Caseload: About Right

Finally, workers were asked to estimate the number of cases they considered to be exceptionally difficult. This was done to understand workers’ perceptions of caseload severity, which was created by dividing the number of difficult cases into the total number of cases. Results are given as a percentage of the total caseload. On average across all waves, workers indicated that about 26 percent of their caseload could be considered very difficult. Over time, scores ranged from 24.6 percent to 28.4 percent.

In addition to asking about the number of cases considered extremely challenging, participants were also asked to provide a short-answer example of a difficult case at each wave of data collection. From an initial review of the short responses at wave 2, nine patterns relating to exceptionally challenging cases were identified: 1. difficulty making contact (e.g., unable to locate client, language barriers) 2. agency-related issues (e.g., lack of supervisory support, high workload) 3. parent/caregiver attitudes or behaviors (e.g., difficult or hostile parents, non-compliant parents) 4. mental health issues 5. substance abuse issues 6. case types (e.g., high-risk cases, cases involving domestic violence) 7. case attributes (e.g., teenagers, multiple children on a single case; medical issues) 8. case tasks (e.g., removals, safety planning, transportation) 9. procedural/paperwork issues (e.g., interacting with the legal/court system, workflow issues).

These nine categories were then organized into three primary themes: (1) agency policies and procedures; (2) specific case types, attributes, and tasks; and (3) client-related challenges.

18 | P a g e

Florida Institute for Child Welfare

1.

Agency Policies and Procedures

2.

Specific case types, attributes, and tasks

3.

Client-related Challenges

Many participants indicated that cases were only exceptionally challenging because they were situated within a high workload, often with elevated time pressure. Beyond this, procedural issues made cases more difficult for workers, including workflow problems, such as having to wait for someone else to complete a task before moving forward on a case. Others described their lack of familiarity with, or uncertainty about procedures and paperwork to be completed. When difficulties did arise, participants did not always feel supported, by either their supervisors or colleagues.

Cases with particularly egregious maltreatment such as sexual abuse, physical abuse with injury, and child death presented difficulties for workers as did those with concurrent issues, such as domestic violence, juvenile justice involvement, repeat offenders, and mental health and substance abuse problems. Often, participants identified complicated family dynamics, which included both the logistical challenges of working with large families, as well as navigating conflict within the client system, such as disagreements between the parent and caregiver. Though some participants spoke to child-specific attributes of a case such as runaways, medical issues, or sibling abuse, most participants described the challenge of multiple children on a single case, or cases in which the plans were very complex due to the need for numerous services or providers. This was especially true in the context of certain case tasks, such as transportation, visitation, and locating placements, as workers felt the time it takes to do these things interferes with their other responsibilities. Notably, some workers also expressed difficulties with tasks related to safety planning, removals, and shelter cases, with some specifically saying they have difficulty in identifying present danger or distinguishing between impending and present danger.

Being unable to locate clients or otherwise having difficulty in connecting with them due to circumstances (e.g., homelessness, incarceration, out of town placements) or lack of contact information also made cases particularly difficult. Among clients that workers were able to contact, those who were disrespectful, argumentative, and rude to the worker, as well as those having a lack of boundaries (e.g., making inappropriate demands on the workers’ time) were considered exceptionally challenging. At times, workers described how these attitudes and behaviors rose to the level of racism or threatening or violent behaviors toward the worker.

Newly hired child welfare workers reported they considered about 27 percent of their caseloads to be exceptionally challenging, with many of them reporting multiple areas of difficulty. Initial findings suggest that newly hired workers are facing both procedural and emotional frustration in these cases.

To better understand these frustrations, FSPSF researchers will continue examining how workers perceive challenging cases, and how those perceptions may change over time. Some elements of exceptionally challenging cases do not have simple solutions; however, the following recommendations may help support newly hired child welfare workers in managing difficult cases: Implications

•

Consider offering “refresher” sessions for newly hired workers regarding policies, procedures, and paperwork. Many new workers in this sample described challenges related to unwritten expectations or lack of knowledge of required documentation. Ensuring that newly hired workers have clear, written direction may alleviate some of this uncertainty.

•

Consider additional content and/or time on present danger during pre-service training and supervision time. Many workers shared that exceptionally challenging cases are those with present

19 | P a g e

Florida Institute for Child Welfare

danger, though there was much variation in their specific challenges, including identification of present danger and safety planning for present danger. Relatedly, workers shared that distinguishing between impending and present danger is challenging. While this might be related to the need for directed training, it may also reflect a need for support and encouragement around the difficulty of making high pressure decisions, especially within the first few months on the job.

Agencies should provide new workers with sustained mentorship to ensure they have the capacity to respond to this and other challenges on cases their caseloads.

•

Promote effective supervision/mentoring to help workers navigate challenges that are unmalleable (e.g., difficult or hostile parents, cases with domestic violence, cases with present danger).

Supervisory support and mentoring could increase workers’ efficacy in addressing these issues as well as help them debrief any emotional reactions they may be having in working these cases.

Topics would ideally be addressed during regularly scheduled, ongoing mentorship with supervisors about cases and caseload responsibilities.

Data: Work Tasks

Workers were asked about the amount of time spent carrying out their work tasks. They were asked to identify the number of visits and the number of hours they spent driving for their work responsibilities in the previous week. These, too, were stable over time. Across all waves of data collection, workers averaged 9.1 visits per week, which ranged from a low of 7.4 visits to a high of 10 visits. The amount of time spent driving for work in the previous week averaged 10.9 hours over the course of the study and ranged from 9.3 hours per week to 11.4. We did not ask how much driving time was with a child or family from the caseload. However, while some time in the car with a client may be useful time for casework, it is notable that on average, about 25% of a typical 40-hour work week was spent driving.

Finally, workers were asked to identify the number of days in the past month that they came in early, stayed late, or worked on their day off in order to complete their assigned tasks. Overall, the number of extra days spent working declined steadily over the course of the study. At the 6-month data collection, on average, workers put in extra time on 9.6 days in the past month. By the 3.5-year point, that had been reduced to 5.7 days.

Client Violence Workers’ experiences of client-perpetrated violence (CPV) is an under-studied topic. The FSPSF team asked questions about nine different types of client violence at each wave of data collection and completed interviews with a sub-sample of participants on the topic. The nine types of client violence represent a range of incidents from being yelled at to being assaulted with injuries severe enough to warrant a visit to a hospital. The different forms of client violence were categorized into three types: 1) non-physical violence included being yelled at or sworn at, and property damage; 2) threats included general threats without any physical contact, threats of damage to property, and threats with a weapon; and 3) assault, which included four types of physical contact ranging from no harm or injury to an assault requiring an emergency room visit. Participants who provided a “yes” response to any of the nine types of violence were included in the relevant category. Categories were not mutually exclusive in that workers could experience multiple types of client violence. However, the categories have been dichotomized so that we do not have a measure of frequency for any given type of violence.

20 | P a g e

Florida Institute for Child Welfare

Figure 13: Client violence by category

Client Violence by Type and Wave % Yes 80 70 60 50 40

Data: Client Violence by Type

Figure 13 presents the different categories of client violence by wave of data collection.

Non-physical violence is ubiquitous with the vast majority of workers describing these experiences at every wave. Conversely, very few workers described an assault in the past six months (about 4% at each wave) and about 40 percent of workers at any given wave reported being threatened.

Worker Responses to Client Violence

Telephone interviews were completed with 34 workers including 18 who experienced 20 physical assault and 16 who did not. When asked about non-physical violence, workers 10 described clients shouting, cursing, and 0 making non-specific threats as “routine.”

Non-Physical Threats Assault Physical violence was less common than non-physical violence and included W2 W3 W4 W5 W6 W7 W8 behaviors such as clients shoving workers or using objects to throw at or hit workers.

Injuries were rarely reported. Child removal, or threat of removal, often instigated both physical and non-physical violence. Workers reported legitimate verbal threats to their lives and intense discriminatory slurs based on race, sex, and sexual orientation. Therefore, workers reported that experiences of non-physical violence could be as alarming as the physical ones.

30

Workers’ Thoughts About Client Violence

While yelling, cursing, or making non-specific threats were routine occurrences for almost everyone, workers were evenly divided on whether these verbal incidents constituted CPV. Tolerant workers considered family context and perpetrator intent and did not label unintentional actions as violent because they were “incidental and accidental.” Tolerant workers believed that their respectful and collaborative approach in working with families protected them from escalating violence. Alternatively, watchful workers classified verbal incidents as violence and mentioned their fear of “minor” actions such as a raised voice. Watchful workers emphasized that seemingly minor actions can quickly escalate to dangerous situations and viewed minor incidents as blatant forms of disrespect of the worker and the role of the child welfare system.

Characteristics of Violent Incidents

Participants indicated that violence generally occurred during the process of child removal or when removal was imminent. Similarly, incidents perpetrated by children or adolescents typically occurred during placement disruption. Workers were empathetic and understanding of families’ situations, regardless of whether they felt CPV was incidental or disrespectful. Many workers described violence as circumstantial rather than vengeful. Notably, several workers described incidents in which clients were violent when others were present, supporting workers’ perspective of the spontaneous nature of violent incidents at highly emotional junctures (e.g., removal), when clients may ignore or not realize potential consequences of their actions.

21 | P a g e

Florida Institute for Child Welfare

Immediate Aftermath

Agencies typically did not address violent incidents with workers, although in several instances, workers did not report their experiences. However, when workers did report violence, agencies often did not respond to the concerns, such as supervisors making jokes about it or telling workers to “suck it up.”

Cases were rarely transferred to other staff, but when done, were typically due to a decision above the supervisor or outside the agency (e.g., a restraining order). Without case transfers, workers faced clients who had been violent towards them. Several workers felt their agencies let them down in these circumstances. The lack of agency response is consistent with a mentality that client violence was “part of the job,” casting those who did not or could not handle it as too fragile for the profession.

Implications

Understanding workers’ experiences of client violence including its context and aftermath can inform agency practice and policy in the following ways: •

Agency culture: Child welfare agency personnel could benefit from changing an acceptance mentality to a culture of “expected, but not accepted.” Supervisors and administrators can respond to workers’ reports of client violence with empathy and support, emphasizing risk reduction and self-care. Documenting and debriefing violent incidents can provide workers, supervisors, and administrators with information on how to handle cases and minimize the risk of future violence.

•

Manualized protocol: To assist in changing the culture, agencies can create protocols surrounding client violence including clear definitions and procedures to increase worker safety.

For example, frontline workers and supervisors should know what to do in various situations, such as when workers are fearful prior to a visit, or specific de-escalation strategies for when they receive a threat or when a client appears potentially violent. In addition, a manualized protocol should outline agency procedures following a violent episode to support workers and prevent future incidents.

•

Skill-based training: Skill-based training for workers at all levels can help workers interpret protocols and prepare for potentially violent interactions with clients. In this study, workers were well-aware of risk factors for client violence, including perpetrator and situational characteristics. However, they handled violence differently. Trainings can equip workers with effective strategies for handling potentially violent incidents, and reinforce procedures in the aftermath of violence, including documentation.

Organizational Culture Workers in the FSPSF cohort provided insight into the support they received while performing their job responsibilities. Quantitative information of supervisory and co-worker support across all waves of data collection is provided below. In addition, a sub-sample of workers completed telephone interviews addressing their social support needs and experiences.

22 | P a g e

Florida Institute for Child Welfare

Data: Supervisor and Co-Worker Social Support

In general, workers’ social support scores were consistent over time (Figure 14) with co-workers typically providing slightly more support than supervisors. In general, on a scale measuring social support from 0 (no support) to 15 (very high support) workers expressed a moderately high level of social support being provided by both groups, with mean scores averaging around 9 at each wave of data collection. Sample items from the social support scale include: “my [supervisor/co-worker] can be relied on when things get touch at work” and “my [supervisor/co-worker] is willing to listen to work-related problems.”

Figure 14: Social support

Social Support by Wave Maximum Score: 15

15.0 12.0 9.0 6.0 3.0 0.0

W2

W3

W4

Social Support: Co-Worker

W5

W6

W7

W8

Social Support: Supervisor

Supervisor Support and Guidance

Workers who participated in the phone interviews were clear about the importance of supervisors, particularly in their early development as frontline staff. Workers typically considered their current supervisor as “hands on” and encouraging or, conversely, as “empty” and challenging. Approximately 50 percent of workers described encouraging experiences (n = 20) and 33 percent described challenging experiences (n = 12). The remaining workers (n = 6) described mixed experiences typically due to supervisor turnover. Workers voiced similar expectations for supervision and recognized a range of supervisor quality. They felt “lucky” or “shafted” depending on their situations. Workers classified their experiences in four domains: 1) availability and approachability; 2) consistency of information being provided; 3) micromanagement; and 4) support. Workers had similar expectations of their supervisors in each domain regardless of their actual experiences. The sections below outline newly hired workers’ expectations and experiences regarding each domain and suggest ways to promote encouraging supervision.

Availability and Approachable

Although newly hired workers recognized the nature of child welfare work meant fast-paced, timecontingent, and unpredictable work, they expected that supervisors would be available and willing to assist them as needed. Rather than requiring a certain frequency of supervision, workers voiced the need for ad-hoc availability. Workers with encouraging experiences felt supervisors were available “24/7,” while workers with challenging ones felt they were “winging it” on their own.

Consistency of Information Being Provided

Newly hired workers anticipated having many questions and expected their supervisors to provide and teach clear, accurate information. Encouraging supervisors met these expectations while challenging supervisors gave contradictory instructions, assigned tasks without justification, and gave the same explanations over and over despite workers expressing confusion.

Micromanagement

Given their ultimate job responsibility of child safety, newly hired workers almost universally expected “micromanagement.” For them, this meant frequent case consultations, case reviews, and detailed 23 | P a g e

Florida Institute for Child Welfare

feedback. Workers with encouraging experiences felt that micromanagement provided opportunities for them both to learn and ensure child safety while workers with challenging experiences desired more micromanagement and felt vulnerable in their case decisions.

Support

Given the demanding nature of the job, in order to ensure child safety new workers expected their supervisors to provide a supportive, team atmosphere. Workers with encouraging experiences commonly described “a great union” and supervisors who would “go to bat” for them while workers with challenging experiences and little support often felt incompetent and that they were “letting down” their supervisors.

Implications

Taken together, these findings yield four main implications.

1. Supervisors can benefit from understanding new workers’ high reliance on guidance in their early careers as caseworkers. Although supervisors cannot always be available, they can establish specified hours for new caseworkers to check in each day on an as needed basis.

2. Developing appropriate competencies for supervisors to meet newly hired workers’ common expectations for available, knowledgeable, case-oriented, and supportive supervisors can guide training and leadership development efforts. Supervision requires a distinct skillset from frontline child welfare work and providing supervisors with essential knowledge and skills can promote successful transition experiences among early career workers.

3. A checklist to guide decision-making for workers to reference during the transition from preservice training to independent casework can help workers increase their sense of competence and confidence on the job. The checklist can provide workers with step-by-step instructions under various conditions as they handle a variety of cases for the first time.

4. Reduced caseloads for both newly hired workers and their supervisors during the first six months on the job can promote a supportive environment that effectively socializes newly-hired employees. Supervisors can spend time to address workers’ job concerns and apprehensions and workers can spend time to learn and complete their jobs tasks.

WORKER RESPONSES How workers respond to their job and the inherent stresses of child welfare work are among the most well examined areas on workforce turnover and retention. Issues of burnout and stress are robust indicators of turnover along with job satisfaction, secondary traumatic stress and other indicators of psychological well-being. 4 The FSPSF measured several indicators of worker well-being which are presented below. They include scales that measured stress, burnout, secondary traumatic stress, and time pressure, along with perceived mental and physical health status by wave of data collection.

Scale scores measuring satisfaction with supervision, pay, and benefits is also presented by wave.

Sample items from each scale include:

24 | P a g e

Florida Institute for Child Welfare

Stress In the past 30 days, how frequently did you feel...

1. that you were unable to control the important things in your life?

2. confident about your ability to handle your personal problems?

Burnout 1. How often are you physically exhausted?

2. Are you exhausted in the morning at the thought of another day at work?

Secondary Traumatic Stress 1. My heart started pounding when I thought about my work with clients.

2. It seemed as if I was reliving the trauma(s) experienced by my client(s).

Time Pressure 1. I have too much work to do in the amount of time that I have.

2. I don't have enough time to do my job effectively.

Satisfaction with Supervision 1. My supervisor is quite competent in doing his/her job.

2. My supervisor is unfair to me.

Satisfaction with Pay 1. I feel I am being paid a fair amount for the work I do.

2. Raises are too few and far between.

Satisfaction with Benefits 1. I am not satisfied with the benefits I receive.

2. The benefits are as good as most other organizations offer.

Data: Worker Responses to Work Responsibilities

Figure 15 presents findings from scales on time pressure, stress, secondary traumatic stress and burnout over time. All scores were transformed so that they can be directly compared. b Overall, workers indicated high levels of time pressure in their jobs. Time pressure began high and gradually lowered with more experience, although it remained the highest scoring measure of work burden. Also, mean burnout scores began in the mid-range of total scores and were stable over time indicating a moderate and unchanging level of burnout during workers’ tenure on the job. Stress and secondary traumatic stress were also stable over time but had a comparatively lower level. Taken together, it appears that the impact of child welfare work on worker well-being begins immediately after assuming caseload responsibilities and as a group, does not attenuate or accumulate with more experience or exposure.

Data were transformed using the proportion of maximum scaling method (POMS). In this method, participant mean scores were divided by the total possible scale score resulting in a value between 0 and 1. Scores closer to 0 suggest that mean scores were low as a proportion of the possible total while the opposite is true for scores closer to 1. For example, a POMS mean score of .8 would suggest that mean scores were 80% of the total possible score.

b

25 | P a g e

Florida Institute for Child Welfare

The experiences of child welfare workers remained fairly constant over time, although it is likely that those with higher scores left their agency. While we do not have data to test that specific assumption, we do know that the experiences of departed workers improved significantly following departure. At each wave of data collection, departed workers indicated significantly lower burnout, stress, secondary traumatic stress, and time pressure in their new positions when compared to workers who remained in their agencies.

Figure 15: Responses to work responsibilities

Work Responses by Wave 1.0 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0.0

W2

W3 Time Pressure

W4 Stress

W5

W6

Secondary Traumatic Stress

W7

W8 Burnout

Perceived Physical and Emotional Well-being

Additional analyses focused on workers’ perceived physical and emotional well-being. Beginning with the baseline data collection, participants were asked about perceptions of their physical health and mental health (categorized into excellent/very good or not). Workers also responded to two scales – one focusing on current feelings of psychological distress (e.g., feeling so sad that nothing could cheer you up) and a second focusing on current experiences of sleep disturbances (e.g., nearly every night taking two hours or more to fall asleep).

All measures of physical and emotional well-being worsened over the first 12 months of employment with the largest changes seen in the first six months. However, after 12 months, different patterns emerged. For example, compared to baseline, the percentage of people indicating their physical health was excellent or very good was significantly lower for the first 12 months of employment, but then remained about the same at the 18-month period. This pattern was the same for workers’ experiences of sleep disturbances. Psychological distress was significantly higher at 6 months compared to baseline, stayed about the same by 12 months, and then significantly increased again at 18 months. Perception of mental health stands out, as the percentage of respondents indicating excellent or very good mental health worsened steadily at every wave starting at 72.5 percent at wave 1 and decreasing to 46.5 percent by wave 4, a notable 36 percent decline. Finally, although declines in most indicators of physical and mental health appeared to stabilize by 18 months, none returned to baseline levels of functioning.

26 | P a g e

Florida Institute for Child Welfare

Implications

Taken together, there are several implications for agencies to consider that could support worker’s wellbeing. Agencies should have a robust plan to address the physical and psychological sequelae of child welfare casework as physical and emotional consequences begin early in a worker’s tenure.

•

Training curricula and on-boarding processes should include education and skill-development for newly hired workers to create individualized self-care plans for responding to the stresses of child welfare work.

•

While the physical and mental health of all workers is important, the negative changes reported in the first 6-12 months of employment are notable and deserve extra attention. This includes ensuring that supervisors can anticipate and supportively respond to workers’ stress as they transition from pre-service training into independent casework.

•

Employee assistance programs should be visible and proactively encourage worker self-care strategies.

•

Agency leadership should examine how prominently self-care is woven into the organizational culture.

JOB SATISFACTION

Figure 16: Job satisfaction Workers were asked to assess their level of satisfaction with several Job Satisfaction by Wave elements of their jobs including Maximum Score: 20 supervision, salary, and benefits 20.0 (Figure 16). Similar to other areas of 15.0 interest, workers’ job satisfaction was 10.0 generally stable across time. Workers 5.0 were most satisfied with their 0.0 supervisors followed by satisfaction W2 W3 W4 W5 W6 W7 W8 with benefits and salary. There was an Satisfaction: Pay Satisfaction: Benefits increase is salary satisfaction at wave 4 Satisfaction: Supervision of data collection, approximately 18 months following pre-service training, possibly correlating with salary increases following certification or taking on other caseload responsibilities. Although salary consistently was a source of worker dissatisfaction, it did not appear to be a primary driver of departure as it was cited by only 6 percent of those identifying organizational reasons for leaving (see Figure 7).

SCALE VALIDATIONS During the course of this study, several different scales were developed and validated to assess different concepts relevant to child welfare work. Each scale is described below, and sample questions are provided. Descriptions of how the scales may be used in an agency setting are also provided, and the full scales and scoring instruction are available in the appendices (Appendix C).

Child Welfare Provider Stigma Inventory The Child Welfare Provider Stigma Inventory 5 (CWPSI) assess workers’ stigmatizing attitudes toward clients in three different domains 1) attitudes; 2) behaviors; and 3) co-worker influence. Overall, this scale assesses the extent to which workers may stigmatize clients or their circumstances in a way that 27 | P a g e

Florida Institute for Child Welfare

impacts the quality of services. Sample items for the attitudes domain include “although I try not to, sometimes I believe the parents I work with are not capable of change,” and “sometimes I feel like the parents I work with lie to get what they want.” Sample items for the behaviors domain include “sometimes I don’t give my best effort to help clients when I think they won’t follow through,” and “I would be embarrassed if clients heard some of the things I said about them to my co-workers/supervisor.” Finally, sample items from the co-worker influence domain include “when my co-workers make fun of the parents we work with, I find I am more likely to do so too,” and “I find myself describing clients using the same negative language that my co-workers use.” Agencies could use the CWPSI to serve as the basis for continuing education or assessing workplace climate and culture on attitudes towards clients.

Supervision Practice in Human Service Scale The Supervision Practice in Human Service scale 6 (SPHS) assesses a range of supervisory practices, with a specific focus on behaviors unique to human services. This scale measures a continuum of supervisory behaviors including those representing productive supervision to those representing harmful supervision. The scale includes four domains: 1) leadership; 2) inattention; 3) unprofessional modeling; and 4) malpractice. Examples of items comprising the leadership domain include “my supervisor has the experience necessary to effectively train and guide me in my work,” and “I believe my supervisor is a good professional role model.” Sample items from the inattention domain include “my supervisor frequently forgets about, changes, or cancels meetings with me,” and “my supervisor seems distracted or in a rush when I meet with him or her.” Items from the unprofessional model domain include “I have heard my supervisor gossip or share others’ personal information in the office,” and “I have heard my supervisor make inappropriate jokes or comments in the office.” Finally, sample items from the malpractice domain include “I suspect that my supervisor engages in unethical behavior,” and “my supervisor uses bullying or intimidation when interacting with me.” By assessing worker perceptions of their supervision, agencies could use the SPHS to identify training needs for supervisors.

Intimate Partner Violence Responder Collaboration Survey The Intimate Partner Violence Responder Collaboration Survey 7 (IPVRCS) was originally developed to assess the effectiveness of collaborations among professional groups in the community who respond to intimate partner violence situations (e.g., IPV advocates, law enforcement, etc.). Using the FSPSF sample, this instrument was further validated with frontline child welfare workers. Three domains were identified: 1) non-territorialism; 2) leadership; and 3) camaraderie. Sample items from the nonterritorialism domain include “there are ‘turf wars’ between responders,” and “responders ‘put up walls’ when working together.” Items from the leadership domain include “agency leadership is open to responders’ suggestions for improving work with other agencies,” and “agency leadership addresses problems between responders as they arise.” Finally, sample items from the camaraderie domain include “responders want to help each other do their jobs better,” and “responders communicate respectfully with one another.” Using the IPVRCS, communities can get a wholistic assessment of the relationship among the network of those who engage with IPV victims, including those involved in the child welfare system.

28 | P a g e

Florida Institute for Child Welfare

Recommendations

The Florida Study of Professionals for Safe Families is unique in research on the child welfare workforce.

By recruiting workers as they began their pre-service training and following them through the early stages of careers, we were able to identify the impact of child welfare work as it happened. In a variety of different analyses, it became clear that the first year of employment, the transition from training to casework, is a critical time for workers. The findings suggest policy and practice changes at both the state and agency levels.

Pre-service Training

First, several findings from this study point to recommendations for the pre-service training process.

Although there are some specific ideas presented below, at minimum, we would recommend that there be greater communication and coordination between training staff and agency-based staff to create the optimal training and transition experiences for newly hired workers.

1. Carefully review the curriculum and content delivery to ensure that information provided to workers is tailored to their specific role. Complaints were raised by case managers that preservice training was too oriented to investigations to be useful for them in carrying out their unique functions. It should be noted, however, that recruitment for the FSPSF was occurring during transition to a new case management specialty curriculum. At minimum, ensuring trainers are comfortable with the updated curriculum and have adjusted teaching strategies accordingly is needed.

2. Ensure that there are mechanisms for on-going communication to assure consistency between the information provided in the classroom and how individual agencies approach practice with children and families. The de-centralized structure of training may introduce challenges for trainees and a lack of consistency between training content and agency practice introduces ambiguity and uncertainty for workers. Identifying agency-based training coordinators is one example of a strategy to facilitate coordination between training staff and agencies.

3. Identify processes to ensure trainees have productive field experiences during pre-service training. Workers indicated that meaningful experiences included structured schedules and mentors who were prepared and able to demonstrate or explain how classroom content applied in the field. Satisfied trainees were able to “practice� skills which facilitated greater confidence in moving from training to casework. An agency training coordinator could also be responsible for arranging the field experiences of trainees thereby creating a single point of contact to oversee the experience of trainees.

Supervisor Competencies

Beyond a focus on the pre-service training process, examining the competencies identified for supervisors is also important.

1. Create competencies that account for the developmental needs of new workers to ensure supervisors are prepared for the unique needs of new caseworkers. Participants considered supervisors as essential to their success as caseworkers, relying on their guidance and support when beginning their careers. Workers have different supervision needs at different points in their employment trajectories. For example, physical and emotional well-being showed the largest declines during the first six months of employment. Having competencies, training, and

29 | P a g e

Florida Institute for Child Welfare

time allotted for supervisors to have the appropriate skills and availability for new workers will help decrease workers’ anxieties about casework.

Agency On-boarding

Workers indicated that agency-specific information was typically not reviewed during training. Instead, they simultaneously needed to begin work with clients while learning about agency policy and practices.

1. Build in opportunities for learning about agency-specific policies and procedures during the pre-service calendar, after training but prior to assuming caseload responsibilities, or prior to beginning pre-service training would help ease the transition.

Culture of Self-Care

Many of the challenges that child welfare caseworkers face—for example, time pressures, volatile interactions with children and families, and the emotional burden of protecting the safety of children—are endemic to the field. Yet the impact on workers is clear – FSPSF participants reported steady levels of stress, burnout, and declines in physical and emotional well-being. Agency leadership should support the importance of self-care and develop a culture promoting self-care as an integral part of doing child welfare work. Strategies to consider include: 1. Prominently address self-care during pre-service training and on-boarding activities. Workers should be informed of the FSPSF findings on how focusing on physical and emotional health selfcare buffers against some of the emotional challenges of child welfare work.

2. Utilize supportive supervision to develop individualized self-care plans with workers, discuss self-care and identify strategies for engaging in weekly self-care activities.

3. Partner with local businesses to provide free or reduced-costs services for a variety of selfcare activities. Suggestions include, but aren’t limited to, gym memberships, massages, or yoga classes. Agencies should also consider providing “quiet spaces” and time at work for employees to use for brief periods of relaxation, guided mediation, or other reflective activities.

FUTURE DIRECTIONS

Overall, investing in the health of the child welfare workforce is of critical importance to the success of child welfare programs and ultimately, the safety and well-being of children and families. Given the prospective design of the Florida Study of Professionals for Safe Families, we learned that the first year of employment is a particularly vulnerable time, and special focus on policies and programs that strengthen the development of new workers is needed. However, analyses of the FSPSF data are ongoing (See Appendix D for a list of academic publications currently in print or under review). We intend to utilize FSPSF data to support additional research as well as support different statewide initiative through the Florida Institute for Child Welfare.

Future Research

We anticipate a research agenda that focuses on: 1. Examining predictors of burnout over time. We believe it is important to learn if the factors that influence burnout change as workers take on different responsibilities or gain experience in the field.

2. Examining predictors of turnover at each wave of data collection to help determine how stressors resulting in departure change over time.

30 | P a g e

Florida Institute for Child Welfare

3. Examining workers’ descriptions of difficult cases over time in order to assess how experience impacts those descriptions and identify professional development needs as workers grow into their positions.

4. Examining the typical career trajectory of caseworkers who remain in child welfare.

Although 80 percent of workers in the FSPSF left their agencies, about 40 percent of the original cohort remained in the field. Identifying common trajectories may help to understand career pathways.

5. Examining what predicts retention to discover and nourish factors that promote remaining in the field of child welfare.

6. Exploring work and family life balance and self-care activities and how they impact turnover and retention decisions.

Legislative Priorities (SB 1326)

Finally, findings from the Florida Study of Professionals for Safe Families have been leveraged by the Florida Institute for Child Welfare to inform several initiatives from SB 1326, passed during the 2020 Florida legislative session. Though unfunded for FY20-21, our findings can contribute to: 1. Curriculum redesign and professional certifications. For example, the FSPSF data on what cases workers perceive as challenging or client violence could provide examples to be used in problem-based learning scenarios as well as potential certifications where more advanced training is needed.

2. Workforce well-being and enhancement (WWE). The PI and Project Manager are part of the planning team for this initiative, which provides both historical knowledge of FSPSF and its major findings, but also leverages the existing relationships between FSPSF team members and WWE stakeholders (e.g., agency leadership, trainers). We anticipate that FSPSF findings will at least partially inform the primary foci of the WWE sites.

1

DePanfilis, D., & Zlotnik, J. L. (2008). Retention of front-line staff in child welfare: A systematic review of research. Children and Youth Services Review, 30, 995-1008.

2

Estrada, M., Woodcock, A., & Schultz, P. W. (2014). Tailored panel management: A theory-based approach to building and maintaining participant commitment to a longitudinal study. Evaluation Review, 38(1), 3-28. https://doi.org/10.1177/0193841X14524956

3

Hughes, S., & Lay, S. (2012). Direct service workers’ recommendations for child welfare financing and system reform.

Child Welfare League of America. Retrieved from: https://www.childwelfare.gov/topics/management/workforce/caseload/.

4

Wilke, D. J., Radey, M., King, E. A., Spinelli, C., Rakes, S., & Nolan, C. R. (2017). A multi-level conceptual model to examine child welfare worker turnover and retention decisions. Journal of Public Child Welfare, 12, 204-231. http://dx.doi.org/10.1080/15548732.2017.1373722

5

Kennedy, S. C., Spinelli, C., & Wilke, D. J. (2019). Development and validation of the Child Welfare Provider Stigma Inventory. Journal of Social Work. Published online 3/29/19. DOI: 10.1177/1468017319837518

6

Osteen, P., King, E. A., & Wilke, D. J. (Under review). Development and validation of the Supervision Practice in Human Services Scale. Submitted to Human Service Organizations: Management, Leadership, & Governance.

7

Magruder, Lisa N., "Working the Front Lines of Intimate Partner Violence: Responders' Perceptions of Inter-role Collaboration" (2017). Electronic Theses and Dissertations. 1308. https://digitalcommons.du.edu/etd/1308

31 | P a g e

Florida Institute for Child Welfare