← Volatility Index

witharrow.ai/volatility-index/methodology

Methodology

How validated findings are sourced, dated, counted and limited.

Measure

The annual charts report validated policy findings in this release and should not be read as an exhaustive count or a statistical estimate of all policy activity. Lines connect annual release counts for readability; they do not claim continuous measurement between years or establish causal comparisons between countries. One finding represents the lowest independently operative substantive national immigration-policy effect. The 2026 period ends September 19, 2026.

Evidence and dates

Each public finding passed automated exact-claim checks against saved official evidence. The evidence gate covers the fields needed to support the finding: an official source, a distinct policy effect, an operative date, national scope, and supporting text. The operative date is the date the effect took operation, not merely an announcement or publication date. Material target groups, conditions, exceptions, transitions, quantities, fees, thresholds, quotas and expiries are retained when they form part of the validated claim. Optional descriptive fields do not block an otherwise supported record; when absent, they are omitted.

Counting

Separable legal effects are split and inseparable effects are merged. A shared date or source alone does not make two records duplicates. Known unresolved duplicate and counting-unit groups are withheld from public totals. Unchanged renewals and routine continuations do not create a new finding.

Coverage

Five countries (Australia, Canada, New Zealand, the United Kingdom and the United States) across 85 country-year cells from 2010 to 2026. A focused 2020–2026 census and repair pass was completed across all five countries for this release. Clusters of separately operative changes on one date are counted separately when they change different rules, tools or groups.

Limits

Automated source validation is not independent human legal review and is not legal advice. The published counts describe this release's validated findings, not a complete census or a statistical estimate of all activity. Unresolved material is preserved only in internal review materials.

Files

Release: five-country-2010-20260922-v118-evidence-presentation-fix. Data from January 1, 2010 through September 19, 2026. Published September 22, 2026.