Differentially private treatment of survey-weight adjustments

Determine how to account for weighting adjustment steps, including nonresponse adjustments and calibration, in differential-privacy guarantees for small-area estimates.

Background

The empirical applications treat survey weights as fixed design weights and therefore omit privacy effects associated with nonresponse adjustments and calibration. This assumption allows the paper to derive exact sensitivities for the Hájek estimator, but it does not cover the full weighting procedures used in operational surveys.

The paper identifies the treatment of these weighting adjustments as an unresolved differential-privacy problem. Solving it would be necessary for privacy guarantees that accurately reflect the data-dependent construction of survey weights.

References

How to account for these weighting adjustment steps is currently an unsolved problem for DP \citep{drechsler2023differential}.

Differential Privacy Guarantees in Small Area Estimation  (2609.09516 - Das et al., 8 Sep 2026) in Section “Applications,” paragraph introducing the reported privacy parameters