Good Stewardship of Eviction Data
Good stewardship of eviction data is critical for protecting the privacy of sensitive information, as well as ensuring that data collection and analysis is adequately addressing the potential for data misinterpretation and misuse. For eviction lawsuits in particular, protecting the privacy of an individual’s data is critical as eviction court records contain sensitive personal information about tenants, including their names and addresses, and this information is commonly used by tenant screening companies and landlords to screen housing applications.
While some courts may have privacy protocols in place when sharing this data with users, it is also the responsibility of data users to implement the proper protections and protocols. Collection and analysis are also critical steps because the complexity and variation in eviction legal processes across court systems, in combination with often poor-quality data, presents an opportunity for misinterpretation.
To help users promote good stewardship, this section offers a high-level checklist to ensure all efforts to collect, store, analyze, and disseminate eviction data are conducted ethically, transparently, and with privacy protection as a top priority.
Checklist for Good Stewardship of Eviction Data
Data Collection: Addressing Data Limitations and Gaps
☐ I understand where the data comes from and who owns the data that I am accessing (e.g., the court or a case management system).
☐ I understand the scope of the court eviction data that I was able to source.
- For example, If I accessed the data through a bulk extract for all civil cases, I would be able to separate out eviction cases from other kinds of civil cases.
- Does the data include residential and commercial evictions? If I’m only interested in residential eviction cases, is there a way to separate out commercial eviction cases?
- Does the data cover eviction cases in which a plaintiff (landlord) is filing a lawsuit only to regain possession of the property, or does it also include monetary damages?
- Does the data include sealed or expunged eviction cases?
Data Storage and Privacy: Protecting the People in My Data
☐ I understand the privacy-protecting protocols in place governing my use of this data, and I have a plan in place for ensuring compliance.
☐ I have access to the necessary data structures for secure warehousing of datasets or data with personal information.
Data Analysis: Ensuring Accurate Conclusions
☐ I understand what each data field in the court data means, and how it relates to the eviction legal process in my jurisdiction.
- If there are gaps in my understanding, I have consulted with and integrated the knowledge of court clerks, civil legal aid providers, or others knowledgeable about the data and eviction legal processes.
☐ I have the necessary court data, in a comprehensive, granular, and high-quality format to undertake the analysis I am interested in.
☐ I am appropriately defining the universe of eviction cases for each analysis and adjusting each metric accordingly (i.e., the numerators and denominators).
☐ Any comparisons to other jurisdictions take into account the highly varied nature of laws and legal processes across jurisdictions.
Data Interpretation and Dissemination: Doing No Harm When Communicating Results
☐ The eviction data analysis that I will share includes a clear description of:
- How the data was obtained;
- The scope of the court data included in the analysis;
- Any limitations of the data and the analysis, including those inherent to all court data (e.g., court data will not include informal evictions, illegal evictions, and de facto evictions that occur outside the court process), and those specific to your dataset (e.g., the data does not include sealed eviction cases filed after a certain date); and
- Definitions of the methods and metrics used to analyze the data and any assumptions made in the process (e.g., eviction cases filed over $10,000 are likely commercial evictions and thus excluded from analysis).
☐ When sharing the results of the analysis, I provide the necessary context to understand why a specific eviction metric is important and advise how it should be interpreted and used.
☐ The data analysis I share publicly is anonymized and summarized so it cannot be linked to any one defendant (tenant). I am communicating the minimum amount of personal information for the analysis to be meaningful but still privacy-protecting.
Note: While this checklist is not exhaustive, it reflects key considerations that users must grapple with and offers a starting place to build off of for good stewardship of eviction data.
Case Study: Privacy and Transparency in Eviction Data-Sharing in Alexandria, Virginia
Type: Data Management Practice
Policy/Action: Responsible Data Governance
Eager to better understand eviction trends impacting residents, local leadership in the City of Alexandria, Virginia, leveraged funding from the American Rescue Plan Act to hire a full-time data analyst to create an eviction database.1 By automating data collection across several sources, including the General District Court’s online case management system, the Alexandria Sheriff’s Office, and local legal aid, the analyst built and maintained a publicly accessible eviction dashboard for the city.
Using an automated computer program, or custom scraper, the analyst collects data from each source weekly, analyzes it, and uploads it to the public dashboard. The dashboard shows trends in key eviction metrics, including eviction filings, writs of eviction (legal notices from a judge or law enforcement to carry out an eviction), percent of cases decided in favor of the landlord, and the average amount of unpaid rent owed by each tenant. The analyst also sends reports on that week’s eviction case details to local legal aid, housing service providers, and other community-based organizations to identify eviction trends, target resources to tenants most at risk of eviction, and evaluate the effectiveness of current interventions.
How does the City of Alexandria ensure that this data—which contains personally identifiable information on a deeply traumatic experience—is being used responsibly and is understood by community partners?
Through a data-sharing agreement with the Sheriff’s Office, the City of Alexandria developed a memorandum of understanding (MOU), reviewed by the City Attorney’s office, that outlined terms and protocols for how the data could be used for each party. To adhere to the MOU, the public dashboard shows only city-wide metrics and no information about individual cases, while legal aid partners govern access to private data that contains identifiable information.
Responsible data stewardship includes transparency about data sources, limitations, and clear definitions for legal processes and eviction terminology. The Alexandria dashboard defines the rates of eviction filings and eviction writs clearly and contextualizes these trends against a timeline of pandemic-related court changes in Virginia courts, including their closure at the start of the pandemic and their observance of eviction moratoria. At the height of the pandemic, the city analyst shared biweekly eviction trends with an eviction prevention task force, aligning data with qualitative information from firsthand accounts of court outreach staff.
Proper data stewardship is an ongoing and iterative process, as this case shows. To be sustainable, it requires maintaining privacy and contextualizing data, devoting human resources, and committing to be responsive to the changing data landscape and audience needs.
Citations
- This position was created within the Office of Performance Analytics within the City of Alexandria’s Department of Community and Human Services.