Governments use alternative data to supplement surveys, censuses and official statistics: linked administrative records can help assess programs, while mobile-location, geospatial, satellite and sensor data can reveal patterns of movement or change. These sources can make some analysis more timely or geographically detailed, but they do not automatically represent everyone or replace established statistics. Their value depends on whether officials can validate the data, use them lawfully and protect people from harm.
What counts as alternative data in government?
“Alternative data” is a broad label, not one standardized statistical category. It can include information already held by public agencies, privately collected records, and observations from devices or remote sensing. Some sources are reused for a new purpose; others are acquired from outside government. Those differences matter: an agency’s authority to link its own records is not the same as permission to obtain or use private location or platform data.
These sources generally complement—not automatically replace—surveys, censuses and official statistics. Surveys can provide information that administrative systems do not record, while a new data stream may offer more frequent updates or finer geographic detail. Combining sources can help answer a specific policy question, but only if their coverage, definitions and quality are understood.
What data do governments use besides surveys and censuses?
| Source | What it can help reveal | Important limitation |
|---|---|---|
| Administrative records | Program participation, service needs and links between government programs and population characteristics. | Records were created for operational purposes, so fields, coverage and definitions may not fit a new statistical question. |
| Mobile-phone location data | Travel, migration, occupancy and other patterns of movement or place use. | Devices and subscribers are not automatically representative of all residents; privacy, legal authority and public trust require attention. |
| Private-sector geospatial data | Mobility, urban change, land use and climate-related analysis. | Access may be commercially restricted, and proprietary data can be difficult to validate, integrate or assess for bias. |
| Satellite imagery, vehicle sensors, video and platform data | Transport and urban planning, including observations of activity and infrastructure. | What can be inferred depends on the data source, its coverage and the specific application; an example or pilot does not establish broad adoption. |
The U.S. Census Bureau’s March 7, 2023 working paper, Use of Mobile Phone Location Data in Official Statistics, reviews statistical and pilot applications including travel and migration patterns, housing-unit occupancy and socioeconomic characteristics. It discusses potential timeliness and coverage benefits alongside legal, ethical, privacy and public-trust concerns. These possibilities should not be read as proof that mobile data represent a whole population.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
#1 Best Overall
- Wiley
- Language: english
- Book - storytelling with data: a data visualization guide for business professionals
How can linked administrative records inform policy?
Agencies can link records they already hold with census or survey information to study who uses a program, estimate service needs or identify where delivery could improve. The U.S. Census Bureau’s page Combining Data – A General Overview, revised March 14, 2025, describes examples including combining Social Security records with Census data to estimate future benefit needs, and combining Medicare, IRS and Census information to estimate children’s health-care needs. It also notes that New Jersey used a Census Bureau tool combining state and federal data in Hurricane Sandy recovery.
These are U.S.-specific examples, not a general grant of access for every agency or jurisdiction. Before linking records, officials need to establish a lawful purpose and authority, determine which records are necessary, and address differences in identifiers, definitions and time periods. The Census Bureau says that data it obtains for linkage are confidential and protected by federal law; linkage is limited to approved research supporting its mission, and public releases are summarized and checked to reduce identification risk. That describes the Bureau’s context, not the legal rules everywhere.
Rank #2
Can mobile-phone data help governments plan transport?
It can provide evidence about movement patterns that may help planners examine routes, demand or connections between places. The result is an additional view of travel, not a complete count of every traveler: people without included devices, people who share devices, and differences in how phones are used can affect what the data show.
A historical example illustrates the potential. In its 2017 overview Big Data in Action for Government, the World Bank described Seoul’s nighttime bus route-planning work using phone call and text data alongside taxi data. The report gives figures of three billion call and text data points and five billion corporate and private taxi data points for that example. Those figures belong to the report’s 2017 account; they are not current totals or evidence that the same service or approach continues today. The report characterized big data as a source of high-frequency, granular information that can offer insights into mobility and economic behavior for policy decisions.
Where does alternative data fit in the policy cycle?
The OECD’s 2019 framework in The Path to Becoming a Data-Driven Public Sector groups public-value uses of data into three activities. A data source may support one or more, but collecting it is not itself a policy outcome.
- Anticipation and planning: Use evidence to design interventions, forecast needs and consider where services or infrastructure may be required.
- Delivery: Apply information to improve implementation, responsiveness or public services, subject to appropriate authority and safeguards.
- Evaluation and monitoring: Track performance, assess impacts and audit decisions. Evaluation needs a credible way to distinguish observed changes from effects attributable to the policy.
The same OECD framework emphasizes governance across government, including leadership, shared rules and standards, interoperable architecture and data infrastructure. Those arrangements help make data usable across agencies without treating every dataset as interchangeable.
Rank #4
How should officials decide whether a source is fit for use?
More frequent or detailed data are not necessarily better data. A practical assessment should start with the decision to be made, then test whether the source can answer it reliably and acceptably. The following criteria synthesize issues raised by the Census Bureau, OECD, NIST and the OECD public-sector framework; they are not a formal government scoring standard.
- Policy relevance and coverage: Does the source measure something connected to the decision, and which people, places or activities are missing?
- Timeliness and granularity: Are updates frequent enough and detail appropriate to the decision, without implying false precision?
- Representativeness: Who is included or excluded, and could participation or device ownership skew the result?
- Accuracy and provenance: How were the data collected, what do fields mean, how stable are they over time, and can the agency validate them?
- Access and continuity: Is there legal authority and a viable procurement or partnership arrangement? Could commercial terms, changing products or discontinued access interrupt the work?
- Interoperability and linkage: Can records be meaningfully combined, and what effort or risk is involved in aligning identifiers, definitions and time periods?
- Privacy, security and disclosure: Could collection, linkage, analysis or publication expose people or sensitive information?
- Transparency and trust: Can the agency explain why it uses the source, how decisions are made and what recourse exists for affected people?
These tests are especially important for private geospatial sources. In its 2022 report Using private sector geospatial data to inform policy, the OECD describes their potential to complement conventional geographic data, but also notes access frameworks, commercial sensitivity, privacy and re-identification risks, integration challenges, and difficulties validating accuracy, integrity, structure and bias. It reports that such difficulties have kept some applications in official statistics at proof-of-concept stage. A promising demonstration should therefore not be mistaken for routine, validated use.
Quick wins for a faster PC:
Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →Best Value
How do governments protect privacy when linking or sharing data?
Privacy protection must be considered throughout the data lifecycle—from collection and processing through analysis and dissemination—not added only when a report is ready to publish. The United Nations Committee of Experts on Big Data and Data Science for Official Statistics’ 2023 UN Guide on Privacy-Enhancing Technologies for Official Statistics discusses methods that can protect data at different stages, including secure multiparty computation, homomorphic encryption, differential privacy, synthetic data, distributed learning, zero-knowledge proofs and trusted execution environments. These methods are not interchangeable, and the guide’s examples include concepts and pilots as well as production implementations.
De-identification also does not guarantee anonymity. NIST’s September 14, 2023 publication, SP 800-188: De-Identifying Government Datasets: Techniques and Governance, advises agencies to define goals and assess disclosure risks before selecting a sharing approach. Options it discusses include publishing de-identified data, publishing synthetic data, offering a query interface that incorporates de-identification, or sharing through a protected nonpublic enclave. Governance measures can include disclosure review boards, measurable performance standards and re-identification studies. NIST cautions that merely masking personal information may not provide adequate de-identification.
The appropriate approach depends on the data, purpose, threat model and who needs access. A public release, a controlled research environment and a system that returns vetted query results create different disclosure risks. Agencies also need to consider security, commercial restrictions and the public’s understanding of how private-sector or linked data are being used.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
Free tools Windows power users keep installed
One-click scans. No signup required.
