Accountability & Electoral Watch

Accountability methodology

MBKRU combines public data, documented commitments, and careful editorial judgement. We do not replace official regulators or electoral bodies.

Press and public affairs — governance accountability in Ghana

How MBKRU documents pledges, scores, and citizen voice — independent civic monitoring, not a government channel or legal finding.

Commitments by MP Browse all commitments Government commitments People's Report Card

Principles

Evidence-led: claims about promises or performance cite sources (manifestos, Hansard-style records, budgets, or on-the-record statements).
Non-partisan process: teams document facts and status changes; political commentary is separate from the dataset.
Corrections: when we get something wrong, we fix the record and note material updates where practical.

Key public terms

These phrases appear in the menu, on data exports, and in the catalogue — use this list to read them consistently.

Commitment (catalogue row): A single editorial record: title, optional narrative, source label, and status. The same record appears in the commitment catalogue, the government-programme lens when tagged, and on a member’s sheet if linked.
Government programme (tag): Editorial flag for rows that surface in the government-programme lens on the commitment catalogue. It is not a government seal of approval — it marks programme- or executive-typed items we follow in the same catalogue.
Parliamentarians tracker: Main navigation label for the Accountability & Electoral Watch hub at /parliament-tracker — roster, live stats, and links into the commitment catalogue. MP pledge sheets and roster browsing live under /promises and share the same active nav state.
Commitment catalogue (Browse all): The single filterable dashboard at /promises/browse — default sitting MPs, optional government-programme lens, filters, CSV, and matching public JSON.
People's Report Card (published cycle): A dated editorial release on the People's Report Card page: MP narratives and scores when MBKRU publishes that batch. Years label publication windows — evidence can stack across Ghana's four-year Parliament toward informed electoral choices. Methodology and roadmap language about scorecards do not replace a live cycle here.
Accountability scorecards (programme language): Pre-election and engagement language used in the roadmap and methodology, including the 90-day run-up. Distinct in UI from a published PRC year until a cycle is released under Report card.

Tracked commitments

Each catalogue row is logged with a title, optional narrative, and a source label (e.g. manifesto section or speech). Status moves through tracking, in progress, fulfilled, broken, or deferred — always tied back to evidence or a transparent rationale for deferral.

Score-style views (People's Report Card)

Report cards summarise multiple dimensions using a published methodology. Scores are explanatory, not legal findings — they complement CHRAJ, the Auditor-General, and the Electoral Commission; they never substitute for them.

People's Report Card — Ghana methodology (v1)

MBKRU does not copy a single foreign scorecard wholesale. We combine international transparency practice with Ghanaian institutions (Parliament, Hansard, constitutional attendance debate, constituency project transparency) and our own editorial governance. Headline scores are published only as editorial products with cited sources — not as Electoral Commission or court findings.

Evidence ladder

Tier A — Official / primary: Parliament materials, cited manifesto excerpts, gazettes, authoritative .gov.gh instructions.
Tier B — Secondary: CHRAJ, Auditor-General, reputable monitors — always labelled and linked.
Tier C — Citizen experience: structured MBKRU Voice and related submissions; reported experience, not adjudicated fact. We use medians and caps in aggregation so coordinated pile-ons cannot dominate the citizen-experience layer.

MP index — three components

Each component is scored and published with its own formula and limitations. A headline combines them with fixed weights (reviewed annually, not changed mid-cycle).

Legislative duty (0–100) — plenary and voting presence (with fair exclusions for Speaker, deputies, documented leave); legislative initiative (bills, motions, questions where records allow); committee and oversight work; integrity sub-score uses proven findings only (no points from raw allegations).
Constituency accountable representation (0–50) — availability (office, surgeries, casework paths); voice for the constituency in Hansard-style records; Ghana-specific constituency project transparency (documented public accounting, not rumour).
Citizen-verified experience (0–30) — triaged structured reports (verified / partially supported / not used for scoring), aligned with democratic-parliament norms on raising concerns, processing input, and feedback loops.

Headline (v1 weights): 0.50 × Legislative + 0.35 × Constituency + 0.15 × Citizen experience

Government performance (parallel)

Programme and manifesto commitments use a commitment lifecycle (not started → in progress → substantially complete → complete / blocked) with evidence packs — similar in spirit to open-government commitment tracking. Citizen reports can illustrate delivery problems but do not alone flip a commitment to “fulfilled.”

Publication governance

Fixed publication cycles and dataset freeze dates.
Pre-publication factual dispute window for Tier A errors (wrong attribution, wrong attendance row); no veto over honest citizen experience summaries except factual mis-identification (wrong MP or constituency).
Published limitations: selection bias, digital divide, coordinated campaigns.

International references we adapted

We drew structure and fairness ideas from the following; Ghana data availability and MBKRU editorial sign-off determine what is live in each cycle:

For diaspora-facing signposting and how accountability tools roll out by phase, see Diaspora support.

Claims, citations & limitations

Catalogue commitments are tracked as editorial catalogue rows with sources (manifesto, speech, official URL) where our QA requires them — not findings by courts or the Electoral Commission.
Report card metrics are explanatory; they complement statutory oversight and never substitute for it.
Voice and situational channels triage citizen input for programme accountability — not legal advice or a government hotline.
Bulk data changes (imports, retags) follow internal dry-run and sign-off before we market them as refreshed public datasets.

When enabled for this deployment, partners can pull the same MP roster, public commitment catalogue, and published report-card cycles through read-only exports. Requests are rate-limited and responses may be cached briefly between refreshes. Partner data & API summarises endpoints, attribution, and fair-use expectations for dashboards and newsrooms.

Whistleblowing guidance ← Parliamentarians tracker (Accountability & Electoral Watch)

Loading…