When Does Late Data Change a Workforce Report?

Research question: How often do controlled late records move a synthetic workforce measure across a predeclared review threshold, and which denominator conditions explain it?

Methodology and scope: Create forty-eight synthetic monthly datasets for headcount, start completion, training completion, and voluntary-turnover examples. Lock each definition and cutoff, then add late numerator, denominator, combined, and unresolved-duplicate patterns.

Protocol and measures: Recalculate original and revised inputs and results. A second reviewer reproduces a stratified sample. Report absolute and relative movement, threshold crossings, reproduction errors, and cannot-determine cases.

Scope boundary: The unit is one synthetic measure for one period under one definition. The protocol studies arithmetic sensitivity, not people, causes, materiality policy, or an external restatement decision.

Inference boundaries: Results can identify sensitive constructed conditions but cannot estimate live error frequency or justify a universal threshold. Reporting, HR, finance, legal, and governance owners decide definitions, disclosure, and restatement.

Limitations: Synthetic populations omit many joins, corrections, privacy constraints, and histories. Findings depend on seed rates and threshold choice. Live study needs a data dictionary, access controls, lineage checks, and approved thresholds.

References: National Institute of Standards and Technology, Cybersecurity Framework 2.0 (https://doi.org/10.6028/NIST.CSWP.29); National Institute of Standards and Technology, Privacy Framework 1.0 (https://doi.org/10.6028/NIST.CSWP.01162020); U.S. Government Accountability Office, Standards for Internal Control in the Federal Government, GAO-14-704G (https://www.gao.gov/products/gao-14-704g); Philippine Statistics Authority, Annual Survey of Philippine Business and Industry (https://psa.gov.ph/statistics/annual-survey-philippine-business-and-industry). These primary publications supply control, privacy, evidence, and sector context. None evaluates Outsourced Employment or the synthetic cases proposed here.