External predictive validity of budget support

Determine whether comparisons supported by Tacet's sequential budget are more likely to replicate or to predict outcomes external to the accounting procedure than comparisons that Tacet refuses.

Background

The evaluation establishes internal properties of Tacet's accounting, such as refusal behavior, affordability, and agreement between static and runtime procedures. It does not test whether those accounting outcomes have external empirical meaning. In particular, the paper does not assess whether a supported comparison is more reproducible or externally predictive than a refused comparison, leaving the practical predictive value of the budget unresolved.

References

Nothing in this paper tests whether support under the budget predicts anything external to the accounting itself, such as whether a supported comparison is more likely to replicate than a refused one. The evidence here bears on whether the accounting is internally sound, and the external question is open.

— Tacet: A Language and Type System for Automatic Statistical Validity Accounting  (2608.27451 - Abuah, 27 Aug 2026) in Section 7, Limitations and Threats to Validity, subsection “Predictive validity”