Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Privacy›Disclosure Risk Assessment
Regression modelDisclosure control

Disclosure Risk Assessment

Statistical Disclosure Risk Assessment · Also known as: Microdata Disclosure Risk, Statistical Disclosure Control Risk Estimation, Istatistiksel Açıklama Riski Değerlendirmesi, Re-identification Risk Assessment

Disclosure Risk Assessment is a probabilistic framework introduced by Duncan and Lambert (1989) for quantifying how likely it is that releasing microdata — individual-level records from surveys or administrative files — will allow an outside party to identify a specific respondent or infer sensitive attributes. It is used by statistical agencies, data custodians, and researchers charged with protecting confidentiality before any public release of person-level datasets.

ScholarGate
  1. Regression model
  2. v1
  3. 1 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Disclosure Risk Assessment
Differential Privacyk-AnonymitySynthetic Data Generation

When to use it

Apply Disclosure Risk Assessment when you must release a microdata file containing personal or sensitive records and need a defensible, quantified measure of re-identification risk prior to publication. Key assumptions are that quasi-identifiers can be enumerated, that population frequency estimates are obtainable, and that the intruder model (what the adversary knows) is explicitly specified. The method is less suited to unstructured data (text, images) or settings where the population universe is unknown. Complement it with utility metrics to balance privacy against analytical usefulness.

Strengths & limitations

Strengths
  • Provides a mathematically grounded, auditable probability estimate of re-identification rather than a qualitative judgment.
  • Works directly with the actual data structure, requiring no additional generative model or synthetic substitute.
  • Scalable to large administrative microdata files when efficient frequency tabulation is used.
  • Output (file-level risk score) is interpretable by non-statisticians and meets regulatory reporting needs.
Limitations
  • Requires credible population frequency estimates; when the sampling frame is unavailable or stratification is complex, estimates can be imprecise.
  • The intruder model must be pre-specified; underestimating adversary knowledge leads to systematically low risk estimates.
  • Focuses on identity disclosure; attribute disclosure (inferring a sensitive value without identifying the person) requires separate modelling.
  • Does not directly prescribe which protection method to apply — it diagnoses risk but does not resolve the privacy-utility trade-off.

Frequently asked

How is Disclosure Risk Assessment different from k-anonymity?

k-anonymity is a deterministic structural rule requiring that every record is indistinguishable from at least k-1 others on quasi-identifiers; it is enforced by transformation. Disclosure Risk Assessment is a probabilistic measurement framework that quantifies re-identification probability given a population model and an explicit intruder assumption. The two are complementary: k-anonymity specifies a protection target; disclosure risk measurement verifies whether that target is met.

What is a 'population unique' and why does it matter?

A population unique is a record whose combination of quasi-identifier values belongs to exactly one individual in the entire target population. Population uniques are the highest-risk records because an adversary who spots that combination in the released file immediately knows which person the record belongs to. Estimating the number of population uniques is therefore the core diagnostic in disclosure risk analysis.

Can the method be applied to continuous variables?

Directly, no: the frequency-counting approach assumes categorical or discretised quasi-identifiers. Continuous variables such as exact income or age in days must first be grouped into intervals before the population cell-size calculation. The choice of interval width is itself a modelling decision that affects both the measured risk and the analytical utility of the released data.

Sources

  1. Duncan, G. T., & Lambert, D. (1989). The risk of disclosure for microdata. Journal of Business & Economic Statistics, 7(2), 207–217. DOI: 10.1080/07350015.1989.10509729 ↗

How to cite this page

ScholarGate. (2026, June 2). Statistical Disclosure Risk Assessment. ScholarGate. https://scholargate.app/en/privacy/disclosure-risk

Related methods

Differential Privacyk-AnonymitySynthetic Data Generation

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Differential PrivacyPrivacy↔ compare
  • k-AnonymityPrivacy↔ compare
  • Synthetic Data GenerationPrivacy↔ compare
Compare side by side →

Similar methods

k-AnonymitySynthetic Data GenerationDifferential PrivacyHistorical Nominal Record LinkageRandomized Response TechniqueDual-Systems Estimation of Irregular MigrationCapture-Recapture for Hidden Crime PopulationsSensitivity Analysis for Unmeasured Confounding

Related reference concepts

De-identification and Privacy-Preserving Data AnalysisGenetic Privacy and Data ProtectionRisk Adjustment and Case-Mix AnalysisExposure Assessment MethodsReproducible ResearchCategorical Data Analysis

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Disclosure Risk Assessment (Statistical Disclosure Risk Assessment). Retrieved 2026-07-21 from https://scholargate.app/en/privacy/disclosure-risk · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
George Duncan & Diane Lambert
Year
1989
Type
Probabilistic risk model
Subfamily
Disclosure control
Scope
Microdata files (individual-level records)
Output
Estimated probability of identity or attribute disclosure
Related methods
Differential Privacyk-AnonymitySynthetic Data Generation
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account