To improve service delivery, anticipate future needs, and prevent potential problems, government agencies increasingly are deploying predictive analytics—the use of historical data to look for patterns and identify trends. Indiana, for example, used predictive analytics across multiple state databases to better understand the correlation between infant mortality and indicators like health care, nutrition, and housing, which enabled caseworkers to make data-driven decisions to address child welfare issues. Chicago used advanced predictive analytics to enhance the process by which health inspectors identified restaurants likely to incur food safety violations. And New York City and Boston turned to predictive analytics to preempt overcrowding in subdivided houses and put measures in place to provide safe, affordable housing.
Los Angeles County’s Child Support Service Department (CSSD) has long been on the forefront of predictive analytics to determine the likelihood of child support payments and develop appropriate interventions for its customers. Its current regression model has allowed the department to forecast child support compliance with fairly high certainty. However, as predictive analytic modeling has advanced and as CSSD has access to more granular economic and demographic data, it is in a position to improve its predictive model and take the department’s decision-making and intervention programs to the next level. CSSD is a data-rich department: it handles a quarter million child support cases a year; employs over 1,400 people, with caseworkers handling on average 600 cases at one time; and between 2014 and 2015 collected nearly $1 billion for LA County families.
With this context in mind, CSSD will partner with FUSE Corps to host an executive-level fellow for one year to augment its predictive analytics model to increase child support payment compliance and the rate of that compliance. An improved model will allow CSSD to better serve its customers—fathers, mothers, and their children—and better fulfill its mission to provide child support services in an efficient, effective, and professional manner. The fellow will have the unique opportunity to advance predictive analytics modeling for child support services in the nation’s most populous and economically complex county.
Project Summary & Potential Deliverables
The following provides a general overview of the proposed fellowship project. This project summary and the potential deliverables that follow will be collaboratively revisited by the host agency, the Fellow, and FUSE staff during the first few months of the project, after which a revised scope of work will be developed and agreed upon by the Fellow and the host agency.
Starting in September 2018, it is proposed that the FUSE Fellow will start updating CSSD’s predictive analytics model for child support compliance to better specify the type of service or enforcement intervention. This will involve determining: whether to continue using IBM SPSS modeler software or explore other open-source data modeling software; which data sets to incorporate in the new model; what predictive analyses the model should offer; how best to train staff on the model; and how to ensure the model will be continuously improved, as data-driven decisions around it evolve.
Success in this project will be evaluated in terms of the fellow’s ability to work collaboratively with CSSD staff to revise the model and increase decision-making and intervention options—not just in predicting the likelihood of child payment compliance but also in recommending steps that can increase payment rates. The model’s success will require deep familiarity with data analysis as well as a high degree of communication, coordination, collaboration, benchmarking, and systems creation for positive, predictive outcomes.
The fellow should have experience in predictive analytics modeling in the private, non-profit or public sector, experience introducing new ideas and systems in complex teams, and the ability to set goals that can be well understood and met through realistic timeframes. The fellow will preferably have familiarity with the social service, finance, or information technology fields.
- Determine status of project and goals and objectives – Meet with CSSD management to become familiar with current predictive analytics model and the software used for data modeling. Work with stakeholders to identify and develop short- and long-term goals with timetables for improving compliance outcomes. Where possible, meet with IT and predictive analytics teams in other LA County agencies to hear lessons learned.
- Conduct landscape analysis, develop a plan and begin modeling – Review and identify additional variables to incorporate into the new model, focusing on economic and financial data. Develop a plan with specific recommendations for improving current data management and analysis. Include key stakeholders in the development of recommendations to ensure representative input and required buy-in. Design and iterate model, ensuring it enables staff to make data-driven decisions about likelihood of child support payment, the type of service or enforcement intervention needed, as well as the complexity of the case.
- Implement an improved predictive analytics model that is sustainable – Implement model, guiding and training CSSD staff to understand the decision making it enables and to conduct validity testing going forward. Help ensure widespread understanding and support by leading internal communication efforts about the model and its evolving role in ensuring compliance objectives are met. Facilitate full documentation of the model and develop effective measurement and evaluation processes for future updates and improvements to it.
- Steven Golightly, Director, LA County Child Support Services Department
- Sara Gaeta-Anguiano, Chief Analytics Officer, LA County Child Support Services Department
- At least 15 years of professional experience in a social service, financial, or business management field, with relevant experience in data analytics, project management and/or modeling.
- Experience in predictive analytics and social service delivery a plus.
- Excellent stakeholder engagement skills and the ability to use facilitative leadership techniques to coordinate stakeholder activities.
- Superior critical thinking and analytical skills, and ability to synthesize complex information into clear and concise recommendations.
- Exceptional written and verbal communication skills with an ease in public presentations.
- Ability to relate to a wide variety of diverse audiences with strong emotional intelligence and empathy.
- Big-picture thinker who can identify opportunities for strategic and empathetic decision making and can introduce a mindset of organizational change.
- Self-motivated, goal-oriented, entrepreneurial leader who can also be an independent worker.
- Persistent in obtaining information and creatively resourceful in identifying solutions to complex problems.
- Flexibility, adaptability, persistence, humility, inclusivity, and sensitivity to cultural and economic differences.
- Understands the need for solutions to support all people in a community regardless of age, disability status, justice-involvement, homeless status, race, religion, gender, immigration status, or ethnicity.
Click the link below to upload a resume and complete the online application questions in lieu of a cover letter. The application process will allow you to indicate interest in more than one fellowship opportunity. You only need to submit one application. Interested candidates are encouraged to apply as soon as possible, as selections will be made on a rolling basis and specific opportunities may close quickly.
A FUSE Fellowship is a 12-month engagement requiring fellows be primarily committed to their partnership with the government agency throughout the year to ensure the success of their project. Fellows are retained as independent contractors of FUSE Corps and are paid an annual stipend of $90,000 through monthly installments. This commitment begins on September 24, 2018 and ends September 20th, 2019.
The FUSE Fellowship is an equal opportunity program with a core value of incorporating diverse perspectives. We strongly encourage candidates from all backgrounds to apply.