UCLA / ACT — Research Data Systems and Applied Analysis for Community Initiatives

Reusable infrastructure and project-specific analysis for statewide community coalitions.

Project Overview

Organization: UCLA / All Children Thrive (ACT)
Context: Statewide community-data technical assistance
Project Period: September 2025–2026
Services Provided: Requirements, information architecture, testing, research-data systems, reusable preparation, maps and applications, analysis, documentation, and stakeholder communication
Systems Delivered: Central ArcGIS Hub, combined-variable exploration, reusable data-preparation workflows, research and interpretation guides, maps, analytical applications, and stakeholder-facing project outputs
Scale: 14 coalitions; five repeat users beyond testing; 35+ personally created products; approximately 17 recurring sources
Verified Results: Recurring data and GIS workflow reduced from approximately four hours to roughly 30 minutes; all 14 coalitions received Hub access, five returned beyond testing, and four users completed the revised end-to-end workflow

Program Context

ACT supported California coalitions working on housing, homelessness, alternative police response, educational policy, environmental conditions, mental-health resources, youth spaces, and other protective factors. Teams generally understood their communities and objectives but did not always have the analytical capacity to locate, evaluate, and translate data for decision-makers.

UCLA’s role was to help determine whether an issue could be demonstrated; at what scale; what context was needed; what data could and could not support; how findings related to objectives; and how evidence should be communicated. Common infrastructure needed to support, not replace, local analysis.

The Initial Operational Need

Delivery was fragmented. Teams repeatedly requested individual variables and basic maps, consuming time better used for interpretation. Existing products were scattered across ArcGIS Enterprise groups, email, and other locations with inconsistent organization. Useful work had accumulated without an effective way for coalitions to find, understand, and build on it.

Central Research and Data Hub

I proposed a centralized ArcGIS Hub combining commonly requested datasets, maps, and applications with contextual and analytical guidance normally repeated during assistance sessions. It was not intended as a larger catalog; it was designed to help teams explore independently and arrive with a better-developed question.

I developed the concept, tested it, and revised it from observed use. All 14 coalitions had access. Five active teams used it beyond testing, and each returned more than once.

Testing and Information Architecture

Five users in the first formal round tried to locate information relevant to their own argument, interpret it, and decide how to use it. Some found useful information but did not know what to do next. Others understood the concept they wanted but lacked the terminology needed to find it. The first architecture began too far downstream.

Users did not experience work as isolated resource types; one observation led to another question, comparison, interpretation, or communication need. I reorganized the Hub around Find → Understand → Analyze → Communicate, connecting exploration to explanations, limitations, comparisons, guidance, and communication resources. Four users subsequently completed the revised end-to-end workflow.

Reusable Data-Preparation Workflows

The public Hub was paired with reusable internal preparation. Suitable sources could be parameterized by year, geography, table, and requested fields. Census retrieval became the fullest example because its stable identifiers, variables, geographies, and conventions allowed direct specification.

The sequence was: Specify the request → retrieve data → identify candidate join fields → validate the join → apply the join → prepare the dataset.

A recurring process involving joins and GIS work fell from approximately four hours to roughly 30 minutes. Instead of manual downloads, multiple interfaces, joins, checks, and configuration, the workflow retrieved required fields, constrained geography, completed joins, and produced a scoped dataset ready for visualization. It standardized technical preparation; defining the research question remained analytical work.

Resource Portfolio

I personally created more than 35 map layers, curated datasets, applications, dashboards, walkthroughs, and research and interpretation guides. Topics included population, economic and community conditions, environment, health, heat, and other public data. Combined exploration, guides, and StoryMap-style resources were reusable; highly specific analysis remained direct assistance.

Resources drew from approximately 17 recurring organizations and datasets—including the Census Bureau, KIDS COUNT, CDC sources, California DataQuest, and environmental-health agencies. This count represents sources, not live integrations.

Changes in Technical-Assistance Use

The combined-variable map became the most-used Hub component because teams could examine multiple dimensions before narrowing an investigation. Basic data-finding requests declined; later requests were more developed and increasingly resembled project-specific analysis.

City Heights began with housing-cost figures, then expanded its exploration to community history, demographic change, collective efficacy, and community bonds. The question moved beyond a static comparison toward change and community context.

Applied Analysis: Stockton Park Prioritization

Stockton was a separate technical-assistance project built around a local decision. A city-aligned process had identified renovation candidates using income, rent burden, environment, green-space access, and other ParkServe priorities. Youth participants did not recognize many as parks they used.

Youth conducted an assessment using desired use, safety, broken facilities, tree-planting opportunities, lighting, access, and other local conditions. The question became: Do youth-selected parks also align with the priority-area framework used in the original selection?

Comparison

I compared youth-selected walksheds with the same ParkServe priority areas. Two of three youth selections had greater overlap than city/CHIP comparison parks. In one representative case, the city-selected park had approximately 73% overlap while the youth-selected park had approximately 95%—about 22 percentage points more.

The finding showed that adding youth knowledge could materially affect which parks appeared aligned with existing priorities.

Stakeholder Application

I translated the comparison into a focused map showing priority areas, walksheds, and overlap percentages. Context layers included youth population, rent burden, community health, and environmental burden without replacing the main comparison. I prepared presentation guidance explaining the result and its use in policy discussion.

The project developed a recommendation to incorporate youth into future renovation decisions through a funded fellowship or similar role. It has not yet been formally presented to the city.

Current Operation and Lifecycle

• A centralized research and data Hub
• A combined-variable exploration environment
• Reusable data and interpretation guides
• Standardized preparation workflows
• More than 35 project and cross-project products
• Direct analysis for project-specific questions
• Stakeholder-facing maps and applications

The Hub was developed for the ACT project period and will retire when the broader project closes. Its reusable methods, preparation processes, and analytical approaches remain applicable to future research and technical-assistance systems.