A recipe for systems change: Predictive modeling and street-level bureaucracy among homeless services

This study analyzes the necessary components to managing a successful systems change by considering the processes of housing people experiencing homelessness across a range of geographic locations. Utilizing Michael Lipsky’s notion of street-level bureaucracy, a nonprofit organization called RE!NSTITUTE™ challenges social service agencies to a “100-Day Challenge™” by granting front-line workers ownership of improving their work free from oversight from their administrators. This paper also gives legitimacy to existing models on systems change by using novel mathematical predictive modeling through using principles of conditional probability to recognize the best “recipe” for systems change. This quantitative approach achieves three key goals. Firstly, it confirms the existence of a hierarchy that suspects transformative change is the most crucial to bring about, followed by relational or mental changes, and imparts a data-driven concreteness to the model, revealing varying shares within each broad level. We innovate alternate quasi-periodic visuals whose impacts are lasting. Next, it offers (even under partial knowledge of the predictors) correlational guarantees that one may use while designing future studies: when changes in relationship, practice, and power dynamics are sure, for instance, changes in mental structures often follow with 50% chance. Finally, we point out, through change-point analysis, how shifts in the rate of reducing homelessness may be attributed to assignable causes such as understaffing, which can hamper successes and workers' ability to maintain consistent attendance. We quantify the impact of those shifts. Several instances of 100-Day Challenges™ across various regions of the United States are analyzed to unearth commonalities. These common findings we offer stress that while variables such as changes in resources (or funding) are often emphasized to achieve social change, changes in practices and relationships (networking) are the most influential ingredients to achieve transformative systems change. The sustainment of workers in their positions is a key driver in maintaining growth in successfully relocating individuals experiencing homelessness.