Independence Score Methodology

Version 0.1 — v0.1 signals ongoing calibration

What the Score Measures

The Independence Score is a 0-100 measure of how well the law, on paper, lets an oversight body operate without undue political influence. It is computed from hard statutory columns on each body's record and is the same formula for every body in every country.

This is a structural independence measure, not an effectiveness measure. A body can have perfect structural independence and still fail to do its job, and a body with weak structural independence can be highly effective under the right leadership. The score is a starting point for inquiry, not a verdict.

Dimensions and Weights

DimensionWeightScoring
Appointment method 20 Independent commission = 20, supermajority legislature = 18, judicial = 16, mixed = 12, legislature = 10, executive = 4
Term length 15 ≥10 years = 15, 7-9 = 12, 5-6 = 9, 3-4 = 5, <3 or at-pleasure = 0
Removal standard 20 Cannot remove before term = 20, impeachment only = 18, for cause = 12, at will = 0
Budget independence 15 Self-funded = 15, fixed by formula = 12, legislative line item = 8, executive discretion = 2
Subpoena + compel testimony 10 Both = 10, one = 5, neither = 0
Records access 5 Full = 5, case-by-case = 3, restricted = 1
Public reports required 5 Yes = 5, no = 0
Prepublication review 10 None = 10, advisory only = 7, legislative = 4, executive = 0
Total100

Score Bands

Strong85-100
Moderate70-84
Limited50-69
Weak25-49
Nominal0-24

What the Score Does NOT Measure

  • Performance. Whether the body actually finds problems and reports them.
  • Resources. Whether the body has enough staff and budget to do the work.
  • Impact. Whether the body's findings are acted upon by the executive or legislature.
  • Political context. Informal pressures, norms, and incentives that can override on-paper independence.

Where external performance ratings exist (INTOSAI SAI PMF, International Ombudsman Institute membership status, GRECO compliance, Transparency International CPI), we cite them as secondary sources on each body's detail page. We do not fold them into our score because the methodologies differ and mixing would obscure both.

Changelog

v0.1 — 2026-04-10
Initial methodology. Covers 4 body types: audit, ombudsman, inspector general, ethics. Wave 1 (US only) is the first live application; weights may be adjusted after reviewing real-world distributions.

Law Enforcement Capability Score (v0.2)

The independence score (v0.1) above measures structural independence — how insulated a body is from the institutions it oversees. It applies to law-enforcement oversight bodies unchanged. But independence does not measure LE-specific capability: a body can be structurally independent and still lack the authority to compel discipline, investigate use of force, see body-cam footage, or include civilians on its own panel.

Methodology v0.2 introduces a separate LE capability score on a 0–40 scale, computed from four axes. The two scores are reported side-by-side and never summed.

The four axes

AxisLevels and points
Discipline authority binding 12 / advisory 6 / none 0
Use-of-force investigation independent 10 / co-investigates 5 / refers 0
Evidence + BWC access full 10 / restricted 4 / none 0
Civilian composition required required 8 / mixed-cap 4 / none 0

Bands

Strong32–40
Moderate24–31
Limited16–23
Weak8–15
Nominal0–7

Diff vs v0.1

  • No removed fields.
  • Seven new indicators: subject domains covered, discipline authority, use-of-force investigation power, evidence access, civilian board composition, and the resulting capability score (plus version tag).
  • One new body classification: purpose-built civilian oversight boards, police commissions, and IPCC-style bodies.

Higher LE capability ≠ “better” oversight. A low LE capability score can mean the LE oversight role is filled by a different body — a strong inspector general, an attorney general, or a court — whose authorities don’t fit this rubric. Read the per-jurisdiction summary alongside the score.