Skip to main content

Governance Compliance Score — Methodology

Alpha Disclaimer

This is an alpha research prototype. AI-generated analysis with human advisory input. Data and model have not been independently verified. Do not use for policy, legal, or administrative decisions. The Lanka Data Foundation accepts no responsibility for decisions based on this analysis.

Overview

The Governance Compliance Score (GCS) evaluates how well Sri Lankan legislative mandates are operationalized through institutional bodies, meetings, and powers. It is based on the Mandate-Action-Outcome triad: legislation creates mandates (M), institutions take actions (A) to implement them, and governance quality determines outcomes (O). The GCS captures the gap between what the law requires and what institutional structures actually deliver.

Composite Formula

GCS = max(0, min(1, 0.30×MCS + 0.35×weighted_avg_IVS + 0.20×weighted_avg_DGQ + PDI)) × DCM

Where MCS = Mandate Completeness Score, IVS = Institutional Vitality Score (body-weighted average), DGQ = Decision Governance Quality (body-weighted average), PDI = Power Dormancy Index (penalty), and DCM = Data Confidence Modifier (0.4–1.0 scaling).

Five Dimensions at a Glance

DimensionWeightDescription
Mandate Completeness (MCS)30%How fully an act defines its institutional mandate: bodies established, powers granted, governance depth, functional breadth, and cross-linkage.
Institutional Vitality (IVS)35%Per-body assessment of meeting clarity, quorum rules, reporting obligations, dissent protection, and leadership structure.
Decision Governance Quality (DGQ)20%Per-body evaluation of composition diversity, member sufficiency, power breadth, and accountability chain.
Power Dormancy Index (PDI)PenaltyPenalty applied when bodies have unknown or superseded operational status, indicating potential legislative deadwood.
Data Confidence Modifier (DCM)ScalingScaling factor (0.4–1.0) based on confidence levels in legislative framework, historical details, and current operational status data.

What Are "Powers"?

Throughout this methodology, powers refers to the specific legal authorities that an act grants to its statutory bodies. Each power is a distinct function the body is authorized (or required) to perform — extracted directly from the legislative text with section citations.

Powers are not abstract categories. They are concrete, enumerable capabilities such as:

"Register and license medicines, medical devices, and borderline products (S.14)" "Inquire into professional misconduct and remove practitioners from register (Section 22D–22F)" "Set Maximum Retail Prices (MRP) for medicines (S.14, S.118)"

Why powers matter to the score

Powers appear in two dimensions of the GCS, measuring different things:

Where usedWhat it measuresHow it's counted
MCS — Total Powers (Dimension 1)Legislative ambition of the entire actSum of all powers across every body the act creates. Normalized against the highest-powered act (27 powers).
DGQ — Power Breadth (Dimension 3)Functional capacity of a single bodyNumber of powers for that specific body. Normalized against the most empowered body (15 powers).

A high Total Powers score means the legislature gave the act a broad governance mandate. A high Power Breadth score means an individual body was designed to be operationally effective, not merely ceremonial.

Concrete example: NMRA Act (2015) — 27 total powers

The NMRA Act creates 6 bodies. Each body has a specific set of enumerated powers:

BodyPowersExamples
National Medicines Regulatory Authority15Register and license medicines (S.14); Set Maximum Retail Prices (S.118); Enforce Good Manufacturing Practices per WHO standards (S.51); Conduct post-marketing surveillance and pharmacovigilance (S.14); Ban or withdraw medicines from use (S.108)
Medicines Evaluation Committee (MEC)4Evaluate medicine applications; Recommend registration decisions; Review safety data; Advise on clinical trials
Medical Devices Evaluation Committee (MDEC)4Evaluate medical device applications; Recommend device registration; Review device safety; Advise on device standards
Borderline Products Evaluation Committee2Classify borderline products; Recommend registration pathway
National Advisory Committee1Advise the Authority on policy matters
Appeals Committee1Hear appeals against Authority decisions
Total: 27

Contrast: Nursing Homes Act (1949) — 3 total powers

BodyPowersExamples
Nursing Homes Advisory Board3Advise Minister on nursing home regulations; Inspect registered nursing homes; Recommend registration decisions
Total: 3

Power distribution across all 18 Health acts

PowersActsExamples
20–272NMRA Act (27), Ayurveda Act (20)
10–195Medical Ordinance (15), NATA Act (12), Private Medical Institutions Act (11), Nurses' Council Act (11), SJGH Board Act (10)
4–93Homoeopathy Act (7), Health Services Act (6), National Health Dev. Fund Act (5)
1–34Medical Wants Ordinance (3), Nursing Homes Act (3), Transplantation of Human Tissues Act (1), Food Act (1)
02Mental Disease Ordinance (0), Poisons, Opium & Drugs Ordinance (0)

The maximum of 27 (NMRA Act) is used as the normalization ceiling for the MCS Total Powers sub-component. The maximum of 15 (NMRA Board, the single most empowered body) is used as the ceiling for the DGQ Power Breadth factor.


Dimension 1: Mandate Completeness (MCS) — Weight: 30%

Mandate completeness measures how thoroughly an act defines its governance architecture. It receives 30% because a well-defined mandate is a necessary foundation, but without operational follow-through (measured by IVS), the mandate alone has limited value.

Sub-components

ComponentWeightDescriptionNormalization
Bodies Count25%Number of distinct statutory bodies established by this act.Max = 6 (the most bodies any single health act creates). Score = raw/6.
Total Powers25%Total number of enumerated powers across all statutory bodies (see What Are "Powers?" above for definition and examples).Max = 27 (NMRA Act — the most powers of any health act). Score = raw/27.
Governance Depth20%Number of governance tiers (e.g., Minister → Board → Committee → Sub-committee).Max = 5 tiers. Score = raw/5.
Functional Breadth20%Number of distinct functional categories (policy, regulatory, advisory, operational, etc.).Max = 6 categories. Score = raw/6.
Cross-Linkage10%Number of explicit cross-references to other acts or institutions.Max = 3 linkages. Score = raw/3.

Column definitions: Raw = actual count, Max = highest value across all acts in the dataset, Normalized = Raw ÷ Max, Contribution = Normalized × Weight.

Weight Rationale

  • Bodies Count & Total Powers (25% each) — They are the primary structural indicators of an act's governance scope.
  • Governance Depth & Functional Breadth (20% each) — They measure governance sophistication but are less decisive than raw structural scope.
  • Cross-Linkage (10%) — Useful for integration but not essential for standalone governance quality.

Dimension 2: Institutional Vitality (IVS) — Weight: 35%

Institutional vitality is the strongest predictor of real-world governance quality. A body that meets regularly, defines quorum, reports to oversight, and protects dissent is more likely to function than one that merely exists on paper. It receives the highest weight at 35%.

Factor Rubrics

Meeting Clarity (Weight: 30%)

Meeting frequency is the strongest single predictor of institutional activity. A body that never meets is effectively dormant.

ScoreLabelCriteria
1.0ClearAct specifies explicit meeting frequency (e.g., "at least once per month")
0.5VagueAct mentions meetings but without frequency (e.g., "from time to time")
0.0NoneNo mention of meeting requirements in the act

Quorum Defined (Weight: 20%)

Quorum ensures collective decision-making rather than unilateral action by a subset of members.

ScoreLabelCriteria
1.0YesAct defines a specific quorum number or fraction for valid decisions
0.0NoNo quorum requirement specified in the act

Reporting Mandated (Weight: 20%)

Reporting creates accountability linkage to oversight authorities, enabling scrutiny and course-correction.

ScoreLabelCriteria
1.0YesAct mandates reporting to a higher authority (e.g., Minister, Parliament)
0.0NoNo reporting obligation specified in the act

Dissent Protected (Weight: 15%)

Dissent protection ensures minority views are preserved, preventing groupthink and enabling future audit.

ScoreLabelCriteria
1.0YesAct allows members to record dissenting opinions or written objections
0.0NoNo provision for dissent recording in the act

Chair Type (Weight: 15%)

Chair independence affects meeting agenda, tone, and willingness to challenge executive positions.

ScoreLabelCriteria
1.0ElectedChair is elected by body members — maximizes independence
0.7Appointed (independent)Chair appointed by Minister but from outside the body
0.5Ex-officioChair is an ex-officio government official — dual loyalties possible
0.0UnknownChair selection not specified in the act

Body Category Weights

IVS and DGQ are computed per-body, then combined as a weighted average using category weights:

CategoryWeightRationale
National Policy Council40%Policy councils set strategic direction and have the broadest governance mandate. Their vitality is most critical to the act's governance effectiveness.
Technical Advisory30%Technical bodies provide specialized expertise. Their dysfunction means decisions lack evidence basis.
Operational Steering20%Operational bodies implement policy. Important but downstream of policy/technical quality.
Administrative Support10%Administrative bodies handle logistics. Least impact on governance quality but still relevant.

Dimension 3: Decision Governance Quality (DGQ) — Weight: 20%

Decision governance quality measures structural fairness — diverse composition, sufficient membership, broad powers, and accountability chains. At 20%, it reflects that good governance structure matters but is less decisive than whether the institution actually operates.

Factor Rubrics

Composition Diversity (Weight: 25%)

Measures how diverse the body's membership composition is (elected, appointed, ex-officio, nominated). Diverse composition prevents capture by a single stakeholder group and improves legitimacy.

ScoreLabelCriteria
1.0HighMultiple appointment pathways (elected + appointed + ex-officio + nominated)
0.7ModerateTwo appointment types (e.g., appointed + ex-officio)
0.4LowSingle appointment type (all appointed or all ex-officio)
0.0NoneComposition not specified in the act

Member Sufficiency (Weight: 15%)

How adequate the body's size is for effective deliberation. Score = min(maxMembers/15, 1.0) — bodies below 15 are penalized. Adequate size enables sub-committees and diverse expertise, but size alone doesn't ensure quality.

ScoreLabelCriteria
1.0Sufficient15+ members — adequate for diverse representation and sub-committees
0.5Marginal7–14 members — functional but limited capacity for specialization
0.0InadequateFewer than 7 members or size not specified

Power Breadth (Weight: 30%)

Proportion of the body's enumerated powers vs maximum across all bodies. Score = powersCount/15, where 15 is the most powers any single body holds (the NMRA Board). Broader powers indicate a body was designed to be effective, not merely ceremonial. This is the highest-weighted DGQ factor. See What Are "Powers?" for the full definition.

ScoreLabelCriteria
1.0Comprehensive15+ distinct enumerated powers covering full governance spectrum
0.5Moderate7–14 powers — adequate operational coverage
0.0NarrowFewer than 3 powers or powers not enumerated

Worked example — Three bodies from the NMRA Act:

BodyPowersScoreCalculation
NMRA Board151.00015 ÷ 15 = 1.0 (Comprehensive — full regulatory, licensing, enforcement, pricing, and rulemaking powers)
MEC40.2674 ÷ 15 = 0.267 (Narrow — limited to evaluation and advisory functions)
Appeals Committee10.0671 ÷ 15 = 0.067 (Narrow — single appellate function)

The NMRA Board's score of 1.0 reflects that it was given the full governance spectrum — from registering medicines to setting prices to banning unsafe products. The Appeals Committee's score of 0.067 reflects its deliberately narrow, single-purpose mandate.

Accountability Chain (Weight: 30%)

Average of three binary indicators: reporting (0/1), audit (0/1), dissent (0/1). Score = sum/3. Accountability mechanisms (reporting, audit, dissent) are the strongest safeguards against dysfunction.

ScoreLabelCriteria
1.0FullAll three: reporting mandated + audit required + dissent protected
0.667PartialTwo of three accountability mechanisms present
0.333WeakOnly one accountability mechanism present
0.0NoneNo accountability mechanisms specified

Dimension 4: Power Dormancy Index (PDI) — Penalty

Power Dormancy is a direct penalty subtracted from the raw score. When a body with legal powers shows no operational evidence, it signals governance failure. The penalty is proportional to the body's category weight, so higher-tier dormant bodies produce larger penalties.

How the Penalty Works

For each statutory body with an unknown or superseded operational status:

  1. A raw dormancy penalty is assessed based on the gap between mandated powers and operational evidence.
  2. The raw penalty is multiplied by the body's category weight (0.4 for National Policy Council down to 0.1 for Administrative Support).
  3. All weighted penalties are summed to produce the total PDI, which is subtracted from the raw score before clamping.

This means a dormant National Policy Council (weight 0.4) produces a penalty 4× larger than a dormant Administrative Support body (weight 0.1).


Dimension 5: Data Confidence Modifier (DCM) — Scaling

Data Confidence scales the final score to reflect how much we trust our source data. If legislative text is clear but operational data is missing, the score is reduced — ensuring that uncertain scores don't appear more reliable than they are.

Three Confidence Dimensions

  1. Legislative Framework — Clarity and completeness of the act's text (sections, schedules, amendments).
  2. Historical Details — Availability of amendment history, gazette references, and parliamentary records.
  3. Current Operational Status — Whether we have verified evidence of each body's current operational state.

Each dimension is rated high, medium, or low.

Modifier Rules

ModifierRuleMeaning
1.0All three dimensions are "high"Full confidence — score reflects reality accurately
0.8Any dimension is "medium" (none "low")Moderate uncertainty — score may be slightly higher/lower
0.6Exactly one dimension is "low"Significant gap — one data source is unreliable
0.4Two or more dimensions are "low"Low confidence — score is largely indicative, not definitive

Final Assembly

The GCS for each act is assembled in six steps:

  1. Compute MCS — Sum of (normalized × weight) across the five sub-components.
  2. Compute weighted IVS — For each body, sum (factor score × factor weight). Then take the category-weighted average across all bodies.
  3. Compute weighted DGQ — Same per-body then weighted-average approach as IVS.
  4. Compute PDI — Sum all body dormancy penalties (negative values).
  5. Combine raw scoreraw = 0.30 × MCS + 0.35 × IVS + 0.20 × DGQ + PDI
  6. Apply DCMGCS = max(0, min(1, raw)) × DCM

Grading Scale

GradeRangeMeaning
A0.70–1.00Strong governance compliance — mandate is well-defined and institutional structures are active.
B0.55–0.69Good compliance — most governance structures in place with minor gaps.
C0.40–0.54Moderate compliance — functional mandate exists but institutional vitality is mixed.
D0.25–0.39Weak compliance — significant gaps in institutional operationalization.
F0.00–0.24Failing compliance — mandate exists largely on paper, institutions dormant or superseded.

Data Sources

Scores are computed from three pre-existing data files in this research repository:

  • act-anatomy-profiles.json — 18 act profiles with statutory body inventories, functional profiles, governance hierarchies, and data confidence ratings.
  • ministry-health-meetings.json — Meeting governance data for 26 statutory bodies: frequency, quorum, reporting, dissent mechanisms, and chair type.
  • 12 deep-dive analysis JSON files — Detailed AI-generated analysis providing cross-references, amendment histories, and operational status assessments.

Limitations

  • Operational status gaps: Many bodies have "unknown" operational status because verification requires field research beyond legislative text analysis.
  • No actual meeting frequency data: Scores reflect what the law mandates, not whether meetings actually occur at prescribed frequencies.
  • AI-generated classifications: Body category assignments and composition diversity assessments are heuristic-based and may contain errors.
  • Single ministry scope: Currently covers Ministry of Health only. Cross-ministry comparisons are not yet possible.
  • No macro governance indicators: Does not incorporate external indicators like budget execution, audit findings, or citizen satisfaction.

Credits & Version

AI-generated with human advisory input. Based on research by the Lanka Data Foundation.

  • Model version: GCS-v1.0
  • Data version: 1.0.0
  • Generated: 2026-02-23