Validate residualization under natural construct–artifact entanglement

Establish an independently validated real-data case in which residualization is evaluated when the construct and artifact are naturally entangled, and determine whether the adjustment improves or damages construct validity in that regime.

Background

The paper studies construct–artifact entanglement using a controlled Gaussian sweep and a semi-synthetic assignment probe. These analyses show that residualization can reduce artifact predictability while simultaneously degrading construct alignment, and that harmful configurations can pass all pre-adjustment gates.

However, the assignment probe induces collinearity by changing which already-verified code candidates enter the sample, rather than using a naturally entangled real-world dataset. Consequently, the empirical validity boundary remains unverified in natural data, and the paper does not provide an independently validated real-data demonstration of residualization improving construct validity under genuine entanglement.

References

There is no independently validated case here of residualization improving construct validity on real data where construct and artifact are entangled. The intervention audit establishes attenuation of the measured format contrasts in a design where the channel is label-orthogonal by construction; the observational regime supports the slice-restricted scope of Section~\ref{sec:attrdef} and nothing wider; the entangled region is covered by the controlled sweep and the semi-synthetic probe of Section~\ref{sec:entangled}, and by nothing drawn from natural data.

— When Residualization Helps an Audit: Format Effects, Slice Gains, and Their Limits  (2609.24194 - Weon et al., 21 Sep 2026) in Section 6, Section 9, subsection “What we have not established”