voter-segmentation-methodology.md

May 15, 2026 · View on GitHub

🗳️ Voter Segmentation Methodology

📊 Eurobarometer Integration · Structural-Segmentation Fallback · Citation Requirements
🎯 EU Electorate Segments · Eurobarometer Primary Data · Required Fallback Rules · Confidence Tiers

📋 Document Owner: CEO | 📄 Version: 1.0 | 📅 Last Updated: 2026-05-14 (UTC) 🔄 Review Cycle: Quarterly | ⏰ Next Review: 2026-08-14 🏢 Owner: Hack23 AB (Org.nr 5595347807) | 🏷️ Classification: Public

🎯 Purpose

This document governs the production of extended/voter-segmentation.md artifacts. It codifies:

Eurobarometer as primary source — how to integrate current-cycle polling data.
Structural-segmentation fallback — what to do when Eurobarometer data is unavailable (fallback ladder, required labels).
Citation requirements — which specific Eurobarometer series to cite and how.
Confidence rules — how the triangulation step maps to confidence labels for this artifact.

Gap this document closes: The 2026-05-14 methodology-reflection artifacts noted that extended/voter-segmentation.md was based on structural analysis rather than current EU opinion polling. No methodology page covered Eurobarometer-or-fallback integration. This document is that page.

🛡️ ISMS Policy Alignment

Policy	Relevance
Hack23 AI_Policy §3	Evidence and provenance — Eurobarometer citations must be versioned
Information Security Policy §6	Evidence handling — structural-fallback claims must be disclosed and graded
Open Source Policy	Eurobarometer data is EU public-domain; cite the specific report, not generic "Eurobarometer"

1️⃣ Eurobarometer as Primary Source

1.1 Why Eurobarometer is Primary

Eurobarometer surveys are the EU-institutional gold standard for EU citizen attitudes. They are:

Published by the European Commission on behalf of the European Parliament.
Based on ≥ 26,000 interviews across all 27 EU Member States.
Available with per-MS, per-demographic breakdown (age, education, urbanisation).
Graded A1–A2 under the Admiralty Code when cited with a specific wave and question number.

1.2 Eurobarometer Source Hierarchy

Survey type	Publication cadence	Key questions	EP Relevance	Admiralty
Standard Eurobarometer (EB)	2× / year (spring + autumn)	QA1–QA22 (trust, satisfaction), national identity, EU membership	Aggregate EU attitudes	A1
Special Eurobarometer (SEB)	Topic-specific	Parliamentary elections, climate, AI, migration	Deep-dive on EP policy domains	A1
Flash Eurobarometer	Rapid, < 30 days	Event-driven (post-election, crisis)	Breaking/weekly runs	A2
EP Parlemeter	Annual	Specific to EP institution trust and election intention	Electoral artifacts	A1

1.3 Canonical Citation Form

Every Eurobarometer citation must include:

[Eurobarometer Standard/Special/Flash EB NN — SEASON YYYY, Question QXX, p. NN]

Examples:

[EB 101 — Spring 2024, QA3a (EU membership positive), p. 14, A1]
[EP Parlemeter 2024, Q3 (trust in EP), p. 8, A1]
[Flash EB 546 — May 2026, Q2 (migration priority), A2]

Do not write generic [Eurobarometer, A1] — the specific wave and question number are required for reproducibility.

2️⃣ Structural-Segmentation Fallback

When Eurobarometer data is unavailable for the current analysis run (EP MCP does not expose a Eurobarometer endpoint; the agentic workflow cannot fetch the publication), the analyst applies the 3-rung structural fallback below.

2.1 Fallback Ladder

%%{init: {"theme":"dark","themeVariables":{"primaryColor":"#1565C0","primaryTextColor":"#ffffff","primaryBorderColor":"#0A3F7F","lineColor":"#90CAF9","secondaryColor":"#2E7D32","secondaryTextColor":"#ffffff","tertiaryColor":"#FF9800","tertiaryTextColor":"#000000","mainBkg":"#1565C0","secondBkg":"#2E7D32","tertiaryBkg":"#FF9800","noteBkgColor":"#FFC107","noteTextColor":"#000000","errorBkgColor":"#D32F2F","fontFamily":"Inter, Helvetica, Arial, sans-serif"}}}%%
flowchart TD
    Q0{"Current Eurobarometer<br/>data available?"}
    Q0 -- "✅ Yes" --> PRIM["PRIMARY<br/>Use specific EB wave + question<br/>Admiralty A1–A2<br/>Confidence: 🟢 HIGH"]
    Q0 -- "❌ No" --> Q1{"Prior-wave EB<br/>data (≤ 18 months)?"}
    Q1 -- "✅ Yes" --> FALL1["FALLBACK RUNG 1<br/>Prior-wave EB + EP2024 turnout data<br/>Label: structural extension of EB [DATE]<br/>Admiralty B2–B3<br/>Confidence: 🟡 MEDIUM"]
    Q1 -- "❌ No" --> Q2{"EP2024 election<br/>results by MS/group available?"}
    Q2 -- "✅ Yes" --> FALL2["FALLBACK RUNG 2<br/>EP2024 election + Hooghe/Marks framework<br/>Label: structural inference from EP2024 results<br/>Admiralty B3–C2<br/>Confidence: 🟡 MEDIUM"]
    Q2 -- "❌ No" --> FALL3["FALLBACK RUNG 3<br/>Institutional-structure analysis only<br/>Label: structural analysis, no current polling<br/>Admiralty C3–F6<br/>Confidence: 🔴 LOW"]

    style PRIM fill:#2E7D32,color:#fff
    style FALL1 fill:#FF9800,color:#000
    style FALL2 fill:#FF9800,color:#000
    style FALL3 fill:#D32F2F,color:#fff

2.2 Rung Definitions

Primary — Current Eurobarometer wave

Activates when: A Eurobarometer Standard, Special, or Parlemeter report published within the last 12 months is accessible and relevant to the segment claim.

Required citation: Specific EB wave + question number + page reference (see §1.3).

Confidence: 🟢 HIGH (A1).

Fallback Rung 1 — Prior-wave Eurobarometer extended by EP2024 election data

Activates when: No current-cycle Eurobarometer is accessible but a prior wave (≤ 18 months old) is available in the analyst's knowledge base.

Method: Use the prior EB wave for underlying segment structure; extend with EP2024 election turnout data by MS to update directional shifts.

Required disclosure:

[VOTER-SEG-FALLBACK: Rung 1]
Primary data unavailable: No current Eurobarometer wave accessible this run.
Fallback basis: EB [NN — SEASON YYYY] (structural extension) + EP2024 turnout by MS [A1].
Admiralty: B2 (EB) × A1 (EP2024) → B2 composite
Confidence: 🟡 MEDIUM

Confidence: 🟡 MEDIUM maximum.

Fallback Rung 2 — EP2024 election results + Hooghe/Marks framework

Activates when: Neither current nor recent Eurobarometer data is accessible, but EP2024 election results by MS and political group are available from EP MCP tools.

Method: Map EP2024 vote shares and turnout by MS to Liesbet Hooghe / Gary Marks multi-level-governance segmentation framework. Segments: Europhiles/Euroskeptics × North/South/V4/Baltic/Nordic blocs × urban/rural × generation (post-Maastricht vs. pre-Maastricht voters).

Required disclosure:

[VOTER-SEG-FALLBACK: Rung 2]
Primary and Rung-1 data unavailable.
Fallback basis: EP2024 election results by MS [A1] + Hooghe/Marks framework [C2 academic].
Admiralty: B3 (structural inference)
Confidence: 🟡 MEDIUM
NOTE: Individual voting behaviour is unpredictable from structural segmentation alone.

Confidence: 🟡 MEDIUM. The Mermaid quadrant chart's axes must be labelled as "Structural orientation" rather than "Current attitude" when Rung 2 is active.

Fallback Rung 3 — Institutional structure only

Activates when: No Eurobarometer and no EP2024 results by MS are accessible.

Method: Use known formal EP group seat distribution and treaty-based MS representation rules to infer broad segment outlines only (e.g. large MS vs. small MS, Northern vs. Southern budget contributors).

Required disclosure:

[VOTER-SEG-FALLBACK: Rung 3]
All polling and election data unavailable.
Basis: Institutional structure (EP group seats, MS weighting) only.
Admiralty: C3–F6
Confidence: 🔴 LOW
CAUTION: This segment analysis cannot speak to voter attitudes. It describes
formal political representation only, not citizen preferences.

Confidence: 🔴 LOW mandatory. Rung-3 claims must appear only in caveat sections; they must not be the basis of any forward-looking electoral claim.

3️⃣ Required Segments and Their Primary Sources

Every extended/voter-segmentation.md artifact must cover all applicable segments below. Where a segment cannot be populated because no data is available, an explicit "data gap" row is required.

Segment dimension	Primary source	Fallback source	Minimum depth floor
Europhile / Euroskeptic	EB QA1 (EU membership positive/negative)	EP2024 group vote share	≥120 words
Regional bloc (North/South/V4/Baltic/Nordic)	EB per-MS breakdown	EP2024 turnout by MS	≥80 words per bloc
Urban/rural	EB urbanisation breakdown	World Bank rural population (% of total) [A2]	≥80 words
Age cohort (pre-Maastricht / post-Maastricht / Gen Z)	EB age-group breakdowns	EP2024 youth turnout statistics	≥80 words
Education level	EB education breakdowns	Eurostat education attainment statistics	≥60 words
Issue salience (top 3 per segment)	EB QA14 (most important issues facing EU)	Policy-domain inference from EP2024 manifestos	≥60 words per segment
Political-group alignment	EP2024 vote by segment	EB QD1 (party preference)	≥80 words
Swing segment (2029 outlook)	EB trend over last 3 waves	Historical EP-election turnout trends	≥100 words

4️⃣ Citation Requirements

4.1 Mandatory Citations per Artifact

Every extended/voter-segmentation.md artifact must cite at least:

One Eurobarometer wave (current or fallback Rung 1) — or explicitly state the Rung-2/3 fallback.
EP2024 election turnout data (from the European Parliament official election results page at europarl.europa.eu/election-results-2024/ or Eurostat dataset TOUR_DEM_ELTURNOUT — EP MCP does not provide election turnout data).
One World Bank demographic indicator for the urban/rural or age/education dimension (via world-bank-get-social-data tool).
Hooghe/Marks academic citation when Rung 2 is active (Hooghe, L. & Marks, G. (2018). Cleavage theory meets Europe's crises. European Journal of Political Research, 57(1), 109–135).

4.2 In-Text Citation Format

| Segment | Size estimate | Primary source | Admiralty | Confidence |
|---------|--------------|---------------|-----------|-----------|
| Europhile (strong) | ~45 % of EU electorate | EB 101 Spring 2024, QA1a | A1 | 🟢 HIGH |
| Rural Euroskeptic | ~18 % of EU electorate | EB 99 Autumn 2023, QA1b + WB rural pop. | B2 | 🟡 MEDIUM |

4.3 Specific Eurobarometer Questions to Cite

Claim type	Eurobarometer series	Specific question
EU membership support	Standard EB	QA1a "EU membership is a good thing"
EP trust	Standard EB	QA13 "Tend to trust: European Parliament"
Most important EU issues	Standard EB	QA14 "Most important issues facing EU"
Life satisfaction (proxy for economic discontent)	Standard EB	D73
EP election intention	EP Parlemeter	Q1 "Likelihood to vote in EP elections"
EP-related awareness	EP Parlemeter	Q5 "Aware of upcoming EP elections"

5️⃣ Confidence Rules for Voter-Segmentation Claims

Claim type	Primary source	Maximum confidence
Current EU-wide attitudes	Current EB wave (< 12 months)	🟢 HIGH (A1)
MS-level attitudes	EB per-MS breakdown (< 18 months)	🟡 MEDIUM (A2)
EP election intention	EP Parlemeter (< 12 months)	🟡 MEDIUM (A2) — election is inherently uncertain
Rung-1 structural extension	Prior-wave EB + EP2024	🟡 MEDIUM (B2)
Rung-2 structural inference	EP2024 + Hooghe/Marks	🟡 MEDIUM (B3) — maximum for structural inference
Rung-3 institutional only	EP group seats only	🔴 LOW (C3+)
2029 electoral projection	Any source, > 3 years ahead	🔴 LOW mandatory regardless of source

5.1 Mandatory Caveat for Non-Primary Data

When Rung 2 or 3 is active, the artifact's Overview section must include:

"⚠️ This voter-segmentation analysis is based on [Rung 2: structural analysis from EP2024 results / Rung 3: institutional-structure analysis only], not on current EU opinion polling. Specific polling data (e.g. Eurobarometer [SEASON YEAR]) would strengthen this artifact. Confidence is capped at 🟡 MEDIUM [Rung 2] / 🔴 LOW [Rung 3]. Individual voting behaviour cannot be predicted from structural segmentation alone."

6️⃣ Relationship to Other Artifacts

Artifact	Voter-segmentation role
`intelligence/synthesis-summary.md`	Cites swing-segment finding from voter-segmentation as a forward-monitor
`intelligence/scenario-forecast.md`	Uses segment-level swing probabilities as scenario inputs
`intelligence/stakeholder-map.md`	Cross-references segment size for citizen-group power scores
`intelligence/coalition-dynamics.md`	May reference the Europhile/Euroskeptic breakdown as context for group alignment
`intelligence/pestle-analysis.md §Social`	Cites demographic and age-cohort breakdown
`extended/coalition-mathematics.md`	Uses national-delegation segments to model electoral swing scenarios

📋 Quick-Reference Checklist

Eurobarometer cited first: Current-wave EB attempted before any fallback.
Fallback rung declared: If Eurobarometer unavailable, explicit [VOTER-SEG-FALLBACK: Rung N] marker in artifact Overview.
Specific EB wave + question: Never generic "Eurobarometer" — wave number + question code required.
Rung 2/3 caveat in Overview: Mandatory disclaimer paragraph when Rung 2 or 3 is active.
All 8 segments covered: Each with depth floor text or explicit data-gap note.
Confidence correctly labeled: 🟢 for primary; 🟡 for Rung 1/2; 🔴 for Rung 3 and all >3-year projections.
Mermaid quadrant labelled correctly: "Current attitude" for Rung 0/1; "Structural orientation" for Rung 2/3.
World Bank demographic source cited: At least once for urban/rural or age/education dimension.

source-triangulation.md — 4-step fallback ladder (voter-segmentation applies Rungs 1–3 from the same logic)
confidence-calibration.md — unified confidence marker rules
electoral-domain-methodology.md — electoral domain methodology (EP election forecasting rules)
electoral-cycle-methodology.md — multi-cycle EP election analysis
osint-tradecraft-standards.md §2 — Admiralty grading (Eurobarometer is A1)
per-artifact-methodologies.md §voter-segmentation — artifact construction rules

Document Control:

Path: analysis/methodologies/voter-segmentation-methodology.md
Classification: Public
Version: 1.0 — Initial codification of Eurobarometer integration and structural-segmentation fallback rules, addressing the gap identified in 2026-05-14 methodology-reflection artifacts.
ISMS Reference: Hack23/ISMS-PUBLIC AI Policy §3 (Evidence and provenance); Open Source Policy (public-domain data citation requirements)