# UB Researchers Develop Predictive Tools for Housing Eviction Trends

> The University at Buffalo's new early warning system utilizes open-source data to predict housing evictions, a vital tool for human service organizations. This initiative aims to empower communities and enhance readiness against housing instability.

**Source**: google.com | **Published**: 2026-08-11 | **Type**: research

## Key Facts

- Open-source tools can predict eviction trends, aiding resource allocation for human services.
- Eviction projections were 2.6-3.3x higher than actual filings, highlighting policy effectiveness.
- Communities of color faced significant job losses, revealing inequities needing targeted interventions.
- Data accessibility issues hindered crisis response, emphasizing the need for better public data systems.
- Enhanced data skills in organizations can improve decision-making during future crises and emergencies.

## Summary

Researchers at the University at Buffalo (UB) have developed an early warning system aimed at predicting housing evictions, a pressing issue that gained prominence during the COVID-19 pandemic. The study, published on August 9, 2026, in the *Journal of Technology in Human Services*, highlights the challenges faced by human service organizations in accurately forecasting community needs due to limited access to data. This initiative is significant as it addresses a critical gap in resources that can help mitigate the impact of housing instability, particularly in economically vulnerable communities.

The impetus for this research arose from the fear of an “eviction tsunami” during the pandemic, when many households struggled to meet rent obligations due to sudden economic disruptions. Maria Rodriguez, a lead researcher and assistant professor at UB, noted that many community organizations lacked the necessary data to identify households at risk of eviction, as detailed housing information is often locked behind paywalls. The study aimed to explore whether open-source data and publicly available tools could provide a viable alternative for predicting eviction trends.

The research team, which included experts from various disciplines, focused on Bronx County, New York—one of the areas most affected by both the foreclosure crisis and the pandemic. They utilized open-source tools to analyze publicly available datasets on eviction filings, demographics, and employment trends, constructing statistical models to project eviction filings from January 2020 to July 2021. The findings revealed that projected eviction filings were significantly higher than actual reports, suggesting that government interventions, such as eviction moratoriums, played a crucial role in preventing a larger crisis.

The implications of this study extend beyond immediate eviction forecasts. It underscores the necessity for human service organizations to enhance their data capabilities. As Rodriguez pointed out, while not every organization requires a full data science team, having personnel skilled in data management and analysis is essential for making informed decisions during crises. This need for data literacy signals a shift in how organizations will need to operate in the future, particularly as economic conditions fluctuate.

The study also highlights persistent inequities in housing and employment, particularly affecting communities of color, which experienced some of the most significant job losses during the pandemic. By leveraging open-source tools, organizations can gain insights into where these inequities are concentrated, enabling them to allocate resources more effectively and advocate for necessary support in the most affected areas.

Looking ahead, the findings suggest a growing market for data-driven solutions that can aid human service organizations in crisis response. As the demand for accurate forecasting of social issues increases, companies that develop and provide accessible data analytics tools will likely find opportunities for growth. The ability to anticipate and respond to housing instability could become a competitive advantage for organizations dedicated to social impact. This shift may also encourage partnerships between technology firms and social service agencies, fostering innovation in how community needs are addressed. 

In summary, the research from UB not only demonstrates the potential of open-source data in predicting housing instability but also emphasizes the need for enhanced data skills within human service organizations. As the landscape of social services evolves, the integration of data analytics will be crucial for effective crisis management and equitable resource distribution.

## Entities

- **Technologies**: open-source tools, statistical forecasting models
- **People**: Keith Page, Maria Rodriguez, Kenneth Joseph, Jan Voltaire Vergara, Erin Dohler, Amy Wilson, John Phillips, Melissa Villodas
- **Organizations**: University at Buffalo, University of North Carolina at Chapel Hill, University of Minnesota Duluth, George Mason University

## Key Concepts

housing evictions, COVID-19 pandemic, data access, open-source tools, statistical forecasting, community needs, inequities, resource allocation

## Definitions

- **open-source tools**: Software that is freely available for use and modification, allowing organizations to analyze data without costly licenses.
- **statistical forecasting models**: Mathematical models used to predict future events based on historical data.
- **eviction moratorium**: A temporary prohibition on evictions to protect tenants during crises.
- **housing instability**: A condition where individuals or families face the risk of losing their housing.
- **community-based organizations**: Local organizations that provide services and support to residents in a specific community.

## Use Cases

- Predicting eviction filings during crises
- Analyzing demographic data for resource allocation
- Forecasting housing instability
- Identifying inequities in employment trends
- Preparing human service organizations for emerging challenges

## Frequently Asked Questions

**What is the purpose of the early warning system?**

The early warning system aims to help human service organizations predict eviction filings during crises, enabling them to respond more effectively to community needs.

**How did the COVID-19 pandemic influence this study?**

The pandemic highlighted gaps in data access and increased concerns about evictions, prompting researchers to explore how open-source tools could forecast housing instability.

**What are the benefits of using open-source tools?**

Open-source tools provide cost-effective access to data analysis capabilities, allowing organizations to generate insights without the burden of expensive licenses.

**What challenges do community-based organizations face?**

Many community-based organizations struggle with limited resources and access to detailed housing data, making it difficult to assess and respond to housing instability.

**Why is stronger data skills important for human service organizations?**

Stronger data skills enable organizations to effectively utilize open-source tools, leading to better decision-making and resource allocation during crises.

## Links

- [Read on Welcome.AI](https://welcome.ai/content/ub-researchers-develop-predictive-tools-for-housing-eviction-trends)
- [Original source](https://www.google.com/url?rct=j&sa=t&url=https://www.buffalo.edu/innovation/ai-health/news-events.host.html/content/shared/university/news/news-center-releases/2026/08/housing-eviction-public-data-study.detail.html&ct=ga&cd=CAIyGmI5MjM0MzI3NzhlNTYzOGE6Y29tOmVuOlVT&usg=AOvVaw1To_-PVjmprQY1MSb6u63v)

---

Source: Welcome.AI | https://welcome.ai/content/ub-researchers-develop-predictive-tools-for-housing-eviction-trends