media-framing-analysis.md

July 8, 2026 · View on GitHub

📰 Media Framing & Influence-Operations Analysis Template

📊 Narratives · Manipulation Vectors · Frame Lifecycles · Global Audience Orientation
🎯 Entman Functions · Cognitive Vulnerabilities · DISARM TTPs · CIB Signals · RRPA · Counter-Resilience

📋 Document Owner: CEO | 📄 Version: 2.3 | 📅 Last Updated: 2026-05-03 (UTC) 🏢 Owner: Hack23 AB (Org.nr 559534-7807) | 🏷️ Classification: Public

📐 Template Contract — every fill of this template MUST satisfy this row.

Slot Value
Owning methodology per-artifact-methodologies.md
Owning gate check Check 5 (Mermaid) + Check 8 (Family D) — see 05-analysis-gate.md
Required inputs search_anforanden, public media archives
Horizon band per-run (per scripts/horizon-context.ts)
Output family Family D — Electoral & Domain
Aggregation order 5 of 30 in canonical order (see scripts/render-lib/aggregator/order.ts)
Reader Intelligence Guide row generated from media-framing-analysis.md (see scripts/render-lib/aggregator/reader-guide.ts)
Canonical evidence anchor | claim | evidence (dok_id / vote / MP intressent_id / primary-source URL) | retrieved_at | confidence | — every analytical claim row uses this schema.

Cross-reference: README.md §Template ↔ Methodology ↔ Gate-Check Matrix.

Slot	Value
Owning methodology	`per-artifact-methodologies.md`
Owning gate check	Check 5 (Mermaid) + Check 8 (Family D) — see `05-analysis-gate.md`
Required inputs	`search_anforanden`, public media archives
Horizon band	per-run (per `scripts/horizon-context.ts`)
Output family	Family D — Electoral & Domain
Aggregation order	5 of 30 in canonical order (see `scripts/render-lib/aggregator/order.ts`)
Reader Intelligence Guide	row generated from `media-framing-analysis.md` (see `scripts/render-lib/aggregator/reader-guide.ts`)
Canonical evidence anchor	`\| claim \| evidence (dok_id / vote / MP intressent_id / primary-source URL) \| retrieved_at \| confidence \|` — every analytical claim row uses this schema.

🔄 Tradecraft Context

Element	Value
F3EAD Stage	ANALYZE → DISSEMINATE — media environment, manipulation surface, and counter-resilience guidance
PIRs Served	PIR-6 (Election Integrity — narrative manipulation detection), PIR-7 (Democratic Norms — media pluralism), PIR-8 (Foreign Influence — coordinated inauthentic behaviour & state-actor amplification), PIR-9 (Cognitive Security — bias-exploitation surface)
Admiralty Floor	No outlet is "neutral" — Admiralty grade reflects process discipline (corrections policy, source attribution, separation of news/opinion), not ideological neutrality. Every cited outlet must carry an ownership / funding / lean tag from the Source Ecology table. Public-service broadcasters with editorial-board independence + corrections policy + ombudsman (SVT, SR, BBC, ARD, NRK, DR, YLE) [B2] ⚠️ publicly funded; politically appointed boards; audience-demographic skew documented; international news wires with multi-source verification (Reuters, AP, AFP, DPA) [B2]; commercial quality press with corrections policy (DN [Bonnier], SvD [Schibsted], FT, Le Monde, NYT) [B3] ⚠️ owner ideology + audience-demographic skew documented; tabloids (Aftonbladet [Schibsted/LO-aligned], Expressen [Bonnier], Bild, The Sun) [C3]; alt/partisan outlets (Nyheter Idag, Samhällsnytt, Riks, Breitbart-cognates) [D3] with `[partisan-aligned]` tag; named-figure social posts [D2]; anonymous social posts [E5] with `[unconfirmed]`; foreign-state-affiliated outlets (RT, Sputnik, CGTN, PressTV, RIA, TASS) [F-flag — actor-attributed propaganda] with `[state-affiliated]` annotation. An outlet's grade can never be promoted to imply "neutral truth source" — it only constrains how the artifact must caveat its reliance on that outlet.
WEP + ODNI	Frame-momentum claims use WEP (gaining traction / fading / dormant); influence-operation attributions require MODERATE confidence floor; foreign-state attribution requires HIGH confidence + ≥ 3 independent indicators
Source Diversity Floor	P2 (frame-dominance): ≥ 2 outlets from distinct categories; P1 (laundering chain): ≥ 1 source per node (fringe + alt + MSM); P0 (foreign-state attribution): ≥ 3 independent indicators across actor / behaviour / content / degree / effect (ABCDE)
SAT(s) Applied	Outside-In Thinking, Indicators & Signposts, Red Cell (adversary frame planner), ACH (≥ 3 competing hypotheses for any state-attribution claim), Premortem on counter-resilience plan
ICD 203 Standards	1 (source quality), 2 (uncertainties — frame attribution), 5 (customer relevance — counter-framing actionability), 6 (logical argumentation — causal chains), 9 (visual information — frame map, laundering chain, lifecycle curve)
Frameworks Referenced	Entman (1993) framing functions · DISARM Foundation TTP taxonomy · SCOTCH model (Source/Channel/Object/Target/Composition/Hook) · ABCDE (Camille François: Actor-Behaviour-Content-Degree-Effect) · Wardle/Derakhshan information-disorder taxonomy (mis-/dis-/mal-information) · RAND firehose-of-falsehood model · Lakoff conceptual frames · Cialdini influence principles

🌍 Global Audience Orientation

Required in every instance. Swedish political context is opaque to readers in Asia, the Americas, and much of Europe. This section provides the minimum orientation for informed reading, plus multi-dimensional alignment keys that replace the naive left/right binary.

Swedish political context (for international readers):

Proportional-representation parliament (Riksdag, 349 seats); 4 % electoral threshold; 8-party system
Current government: name the coalition, list member parties and their ideological anchors
Next election date and days remaining; coalition arithmetic (seats, majority threshold, pivotal parties)
NATO/EU/Schengen membership context relevant to the current story
Swedish krona (SEK) — not Euro — monetary sovereignty context where economic frames are active

Multi-Dimensional Alignment Key — never characterise outlets with a single left/right score. Use all five axes and populate the Outlet Bias Audit accordingly:

Axis	Description	Swedish anchors
Economic policy	Interventionist (tax/spend) ↔ Market-liberal (cut/privatise) ↔ Social-democratic (universal services)	S/V/MP: interventionist; M/C/L: market-liberal; SD: economic-nationalist; KD: social-conservative
Social / identity	Progressive (open migration, pluralism) ↔ Conservative-nationalist (border control, national culture)	V/MP: progressive; S: centre-progressive; C/L: liberal-centre; M/KD: conservative-centre; SD: nationalist
EU integration	Federalist (more EU) ↔ Cooperative (national interest within EU) ↔ Sovereigntist (less EU, exit risk)	MP: federalist; S/M/C/L/KD: cooperative; V: critical-cooperative; SD: Eurosceptic-sovereigntist
Security / defence	NATO hawkish (deterrence-first) ↔ Nordic cooperative (soft-security emphasis)	All parties post-2024: broadly NATO-cooperative; V/MP: more Nordic-cooperative; M/KD/SD: hawkish
Media ownership	Public-funded (licence-fee + state boards) ↔ Commercial (ad/subscriber) ↔ Owner-ideological ↔ Partisan-aligned	SVT/SR: public-funded; DN/Expressen: Bonnier commercial; Aftonbladet: Schibsted+LO; Nyheter Idag/Riks: partisan-SD

Regional reader notes:

East Asia (Japan, Korea, China): Sweden as Nordic social-model reference; Sweden's NATO accession 2024 is geopolitically salient; demographic/immigration debates mirror domestic Japanese low-immigration consensus; China tracks Sweden as NATO frontline asset
South/Southeast Asia: Sweden is primarily read as diaspora remittance source and social-model reference; migration debate affects diaspora communities directly
Americas (North): US MAGA frame as comparative prototype for right-populist frames; US tariff shock directly impacts Swedish manufacturing (Volvo, SSAB, Ericsson, SKF); Canadian immigration debate parallel
Americas (South/Central): Sweden as Nordic-model reference for left-progressive parties; migration narrative contrasts with Latin American emigration framing
Europe (EU): Migration policy critical to EU Pact on Migration and Asylum (EUAA 2024); ECHR compliance watched by Council of Europe; Nordic neighbours (DK, NO, FI) use Sweden as comparative baseline
Middle East / Africa: Swedish migration decisions directly affect asylum-seeker flows; ECHR compliance is stakeholder-relevant for source countries

📋 Framing Context

Field	Value
Framing ID	`FRM-YYYY-MM-DD-NNN`
Generated	`YYYY-MM-DD HH:MM UTC`
Subject	`e.g., HD03262 permanent-residence reform package`
Coverage window	`2026-04-28 to 2026-05-01`
Horizon band	`[72h / week / month / quarter / year / cycle / election]`
Salience tier	`High (full template) / Low (baseline update — dominant frame + active influence-operation signals only)`
Save path	`analysis/daily/${ARTICLE_DATE}/${DOC_TYPE}/media-framing-analysis.md`
Source scope	Public media + public social-media posts only — no paywalled scraping, no private-account content, no leaked or hacked material
Outlets reviewed (Sweden)	`DN, SvD, Aftonbladet, Expressen, SVT, SR, TV4, Dagens ETC, Kvartal, Nyheter Idag, Samhällsnytt, Riks, regional press sample`
International outlets reviewed	`Reuters / AP / DPA / AFP / Politico EU / FT / NYT / Le Monde / Der Spiegel / Helsingin Sanomat / Aftenposten / Berlingske / South China Morning Post / NHK World / Al Jazeera English`
State-affiliated outlets monitored	`RT, Sputnik, RIA, TASS, CGTN, Global Times, PressTV` (amplification fingerprinting only — never cited as factual sources)
Counts	`N articles / N broadcast segments / N editorials / N social posts (public)`
Overall Confidence	`🟧 MEDIUM [B2]`

🧭 Frame Package Overview

graph LR
    STORY["📰 Lead story (e.g., HD03262)"] --> F1["🟢 Frame A<br/>government-aligned"]
    STORY --> F2["🔴 Frame B<br/>opposition-aligned"]
    STORY --> F3["🟡 Frame C<br/>establishment / centrist-consensus<br/>(NOT 'neutral' — Bonnier/Schibsted owner ideology,<br/>expert-establishment recruitment, centrist-urban audience)"]
    STORY --> F4["🟣 Frame D<br/>public-broadcaster proceduralist<br/>(public-broadcaster framing — politically-appointed boards,<br/>institutional editorial culture)"]
    STORY --> F5["⚫ Frame E<br/>foreign / state-affiliated overlay"]

    F1 --> OUT1["Outlets: SvD, TV4, regional press, M/SD/KD/L press offices"]
    F2 --> OUT2["Outlets: Aftonbladet, Dagens ETC, S/V/MP press offices"]
    F3 --> OUT3["Outlets: DN, Kvartal, FT, Politico, expert panels"]
    F4 --> OUT4["Outlets: SVT, SR Ekot, parliamentary correspondents"]
    F5 --> OUT5["Outlets: RT/Sputnik/CGTN ⚠️ [state-affiliated], cognate fringe networks"]

    style STORY fill:#1565C0,color:#FFFFFF
    style F1 fill:#4CAF50,color:#FFFFFF
    style F2 fill:#D32F2F,color:#FFFFFF
    style F3 fill:#FFC107,color:#000000
    style F4 fill:#7B1FA2,color:#FFFFFF
    style F5 fill:#212121,color:#FFFFFF

Frame E inclusion rule: include ONLY when ≥ 1 state-affiliated or coordinated foreign-amplification signal is observed in the coverage window. Otherwise omit Frame E and document the absence explicitly: Frame E: [no foreign-amplification signal observed in window].

Frame C / D labelling discipline: Frame C is the establishment / centrist-consensus frame — never written as "neutral" or "balanced" because it carries the ideology of concentrated commercial owners (Bonnier ≈ 50 % of Swedish national daily-press circulation; Schibsted controls Aftonbladet + SvD), expert-establishment recruitment networks (Stockholm-school-of-economics / Uppsala / Lund alumni clusters), and a centrist-urban audience demographic. Frame D is the public-broadcaster proceduralist frame — never written as "impartial" because SVT/SR/UR boards are appointed via Förvaltningsstiftelsen with parliamentary input, funding is set by government appropriation, and editorial culture is shaped by public-service remit (opartiskhet och saklighet) interpretation that itself is contested. Both frames are positions, not absences of position.

🗂️ Frame Package Table — with Entman functions

Frame	Problem definition	Causal attribution	Moral evaluation	Treatment recommendation	Lead messengers	Approx. share
🟢 A — `[name]`	`what is broken`	`who/what caused it`	`why it is wrong/right`	`what to do`	`Named MP + party + role`	`XX %`
🔴 B — `[name]`	…	…	…	…	…	`XX %`
🟡 C — `[name]`	…	…	…	…	…	`XX %`
🟣 D — `[name]`	…	…	…	…	…	`XX %`
⚫ E — `[foreign overlay if observed]`	…	…	…	…	…	`XX %`

Source: Entman, R.M. (1993). "Framing: Toward Clarification of a Fractured Paradigm." Journal of Communication 43(4): 51–58. Every cell must trace to a dated quote, headline, or dok_id; vague summaries are rejected at Pass 2.

🧠 Cognitive Vulnerability Map

Frame	Bias exploited	Mechanism	Inoculation lever
🟢 A	Loss aversion + status-quo bias	Framing reform as protection of existing entitlements	Reframe loss-language as opportunity-language; cite distributional data
🔴 B	In-group / out-group + scapegoating	Personifying harm onto a single actor or class	Counter-stereotype exemplars; structural-cause data
🟡 C	Authority bias	"Experts say…" without expert disclosure / "centrist-consensus" presented as objective	Expert-credential + funding-source transparency; competing-expert panel; ownership disclosure of the outlet running the "experts say" frame
🟣 D	Availability heuristic + recency + authority bias	Coalition rumour amplified by recent procedural event; public-broadcaster procedural framing read as "what really happened" when it is itself an editorial choice	Base-rate data on coalition stability; remind readers that public-broadcaster framing reflects an institutional editorial culture, not absence of one
⚫ E	Repetition / illusory truth + reactance	Identical phrasing across coordinated nodes	Prebunking of frame-template; show coordination evidence

Source: Cialdini (2001) Influence; Kahneman (2011) Thinking, Fast and Slow; Roozenbeek & van der Linden (2019) "Fake news game confers psychological resistance against online misinformation." Palgrave Communications. Every row must cite a documented bias with primary literature, not folk-psychology claims.

🛰️ Manipulation Indicators — DISARM TTP Map

DISARM Foundation is the open-source MITRE-style taxonomy for cognitive-security threat behaviours (https://www.disarm.foundation/). Use the T#### codes verbatim. Mark [unconfirmed] if any indicator lacks ≥ 2 independent corroborations.

DISARM TTP	Observation in window	Evidence (URL / dok_id / public account)	Confidence	ABCDE attribution
`T0049 Flooding`	High-volume identical posts in <2 h window on `#hashtag`	`[link or "no signal"]`	LOW / MOD / HIGH	Actor: `[unattributed]`; Behaviour: coordinated; Content: identical; Degree: N posts; Effect: trending
`T0086 Astroturfing`	Mass-identical "spontaneous citizen" letters to editor	`[link]`	…	…
`T0040 Demand insincere apology`	Manufactured outrage cycle	`[link]`	…	…
`T0091 Sell merchandise`	Branded merchandise driving frame reach	`[link]`	…	…
`T0099 Prepare assets impersonating legitimate entities`	Spoof accounts of named outlets/MPs	`[link]`	…	…
`T0023 Distort facts`	Doctored quote / out-of-context clip	`[link]`	…	…
`T0085 Develop AI-generated text`	LLM fingerprints in identical posts	`[link]`	…	…
`T0088 Develop AI-generated images / deepfakes`	Synthetic media observed	`[link]`	…	…
`T0118 Amplify existing narrative`	State-affiliated outlet picks up domestic frame	`[link]`	…	…

No-signal finding is also a finding: if no DISARM TTP is observed, write No coordinated manipulation signal in window — frames appear organically driven by public actors. Silence is not the same as "we did not look."

🔗 Narrative-Laundering Chain

Track how a frame migrates from the fringe layer to mainstream coverage. If the chain is incomplete, document the missing link rather than infer it.

flowchart LR
    F[🪨 Fringe<br/>chans, niche Telegram/X accounts<br/>0–10 k reach] --> A[🌗 Alt-media<br/>Nyheter Idag, Samhällsnytt,<br/>partisan podcasts<br/>10 k–100 k]
    A --> P[🟪 Politician amplification<br/>named MP / party-aligned account<br/>>100 k]
    P --> M[🟦 Mainstream<br/>DN/SvD/SVT/SR pickup<br/>1 M+]
    M --> I[🌐 International<br/>Reuters/Politico EU/FT pickup]

    classDef fringe fill:#212121,color:#FFFFFF
    classDef alt fill:#37474F,color:#FFFFFF
    classDef pol fill:#7B1FA2,color:#FFFFFF
    classDef main fill:#1565C0,color:#FFFFFF
    classDef intl fill:#0277BD,color:#FFFFFF
    class F fringe
    class A alt
    class P pol
    class M main
    class I intl

Stage	First observed (UTC)	Carrier	Reach (est., public metrics only)	Admiralty	Notes
Fringe	`YYYY-MM-DD HH:MM`	`[anonymised channel name OR "no fringe origin observed"]`	`~N`	E5	`[unconfirmed]` if only 1 trace
Alt-media	…	`Nyheter Idag / etc.`	…	D3	…
Politician amplification	…	`Named MP (party)`	…	B2/D2	quote URL + dok_id where available
Mainstream	…	`DN / SVT / etc.`	…	B2/C3	…
International	…	`Reuters / Politico EU / FT`	…	B2	…

🌐 Source Ecology — Outlet Bias Audit (no outlet is neutral)

Doctrine: every outlet appearing anywhere in this artifact MUST appear in this table. No bias audit = no citation. Columns capture structural bias drivers (ownership, funding, board, audience), not journalist personalities. Use the multi-dimensional lean columns matching the 5-axis framework in the Global Audience Orientation section. Never use a single left/right score. Sources: Nordicom Media Ownership Database (nordicom.gu.se), Reuters Institute Digital News Report 2024, Allmänhetens Pressombudsman / PO/PON complaint registry, Förvaltningsstiftelsen (SVT/SR/UR board appointments), EUvsDisinfo case dossiers, EU DSA transparency reports, public SÄPO/MUST/FRA/EU-EEAS statements.

Outlet	Ownership group	Funding mix	Board-appointment	Economic axis	Social/identity axis	EU axis	Security axis	Media-ownership axis	Trust %	PO/PON (12 mo)	Foreign-actor link
`DN`	Bonnier News (Bonnier family)	~90 % commercial	Bonnier-appointed	Market-liberal	Centre-progressive	Pro-EU cooperative	Cooperative	Commercial / owner-ideological (Bonnier)	`XX %`	`N`	None
`SvD`	Schibsted Media (Norwegian-listed)	~92 % commercial	Schibsted-appointed	Centre market-liberal	Conservative-centre	Pro-EU cooperative	Cooperative	Commercial / owner-ideological (Schibsted)	`XX %`	`N`	Norwegian public equity
`Aftonbladet`	Schibsted (88 %) + LO (9 %)	~97 % commercial + LO	Schibsted + LO observer	Social-democratic	Progressive	Pro-EU	Cooperative	Commercial + labour-movement (Schibsted + LO)	`XX %`	`N`	None
`Expressen`	Bonnier News	~98 % commercial	Bonnier-appointed	Market-liberal populist	Liberal-right	Pro-EU	Cooperative	Commercial / owner-ideological (Bonnier)	`XX %`	`N`	None
`SVT`	Förvaltningsstiftelsen (public)	~95 % licence-fee	Parl.-influenced boards	Public-service remit (proceduralist)	Proceduralist	Proceduralist	Proceduralist	Public-funded / politically-appointed boards	`XX %`	`N`	None
`SR`	Förvaltningsstiftelsen (public)	~95 % licence-fee	Parl.-influenced boards	Public-service remit (proceduralist)	Proceduralist	Proceduralist	Proceduralist	Public-funded / politically-appointed boards	`XX %`	`N`	None
`TV4`	Allente (Telenor + NOR state minority)	~99 % commercial	Allente-appointed	Commercial centrist	Commercial centrist	Cooperative	Cooperative	Commercial / foreign-state minority (NOR via Telenor)	`XX %`	`N`	Norwegian state minority via Telenor
`Dagens ETC`	Co-operative + reader-owned	~55 % subscriber	Co-operative members	Interventionist	Progressive	Federalist	Nordic cooperative	Co-operative / reader-owned	`XX %`	`0`	None
`Kvartal`	Private (founder + donors)	~70 % subscriber	Founder-controlled	Market-liberal contrarian	Conservative intellectual	Cooperative-critical	Cooperative	Private / donor-funded	`XX %`	`0`	None
`Nyheter Idag / Samhällsnytt / Riks`	Private partisan-aligned	Advertising + donor	Founder-controlled	Economic-nationalist	Conservative-nationalist	Eurosceptic	Hawkish-nationalist	Partisan-aligned / donor-funded	<5 %	Multiple	Document RT/CGTN cross-links
`Reuters / AP / AFP / DPA`	Wire services (various non-profit / private)	Subscription-licensing	Independent boards	Process-disciplined wire	Process-disciplined	Centrist-international	Centrist-international	Non-profit co-op / private / subscription	65–70 %	0	None
`RT / Sputnik / CGTN / PressTV`	RU / CN / IR state-controlled	100 % state	State-appointed	Foreign-state propaganda `[state-affiliated]`	Foreign-state propaganda	Sovereigntist	State-strategic	State-controlled / 100 % state-funded	n/a	n/a	YES — actor-attributed
`[other cited outlet]`	…	…	…	…	…	…	…	…	…	…	…

No-claim-of-neutrality rule: any cell that reads "neutral", "impartial", "balanced", "objective" is rejected at Pass 2. Use the multi-dimensional lean descriptors above. The no-claim-of-neutrality rule applies in the artifact body too — every reference to a cited outlet must carry its multi-dimensional lean tag.

🤝 Coordinated-Inauthentic-Behaviour (CIB) Signal Block

Apply the ABCDE framework (Camille François, Actors, Behaviours, Content, Degree, Effects of Disinformation, 2020) and the Meta CIB removals taxonomy.

Indicator	Observation	Threshold	Status
Account-creation burst	N new accounts posting frame in <72 h	> 5 % of frame's amplifier set created in <72 h	🟢 / 🟧 / 🟥
Posting-time clustering	Posts concentrated in a narrow UTC window incompatible with claimed location	> 3σ from baseline	…
Cross-platform identical phrasing	Verbatim copy across X / Telegram / TikTok / Reddit	≥ 3 platforms, ≥ 5 instances	…
Spoof / impersonation accounts	Look-alike usernames of named MPs or outlets	≥ 1 confirmed spoof	…
Hashtag co-occurrence anomaly	Hashtag pair rises >10× over baseline	EUvsDisinfo / Meta CIB report	…
Bot-likelihood score (Botometer-class)	Median > 0.7 across amplifier set	published academic threshold	…
Fake-engagement spike	Likes/RT > expected by impressions	platform transparency report	…

Attribution discipline: observing CIB ≠ attributing to a state. State attribution requires the public-record floor in the Source Ecology table above.

📡 Algorithmic-Amplification Asymmetry

Doctrine: there is no neutral algorithm. Every recommender / feed / "For You" optimises an objective function (engagement, watch-time, retention, ad-yield) and that function has documented partisan asymmetries. Citing "the algorithm shows X" without naming the optimisation target and the academic/transparency-report measurement is rejected at Pass 2.

Platform	Frame A reach	Frame B reach	Asymmetry ratio	Optimisation target & documented partisan/emotional asymmetry (cite source)
X / Twitter	…	…	A/B	Engagement-maximising algorithmic feed; documented right-amplification asymmetry (Huszár et al. 2022, PNAS 119/1) — no neutral baseline
Facebook	…	…	…	Engagement + group-affinity; News-in-Feed deprecation (Meta DSA 2024); group-amplification still active and documented to amplify outrage and out-group hostility (Rathje, Van Bavel & van der Linden 2021, PNAS 118/26)
TikTok	…	…	…	"For You" — opaque retention-maximising; emotional-content amplification documented (TikTok DSA transparency report 2024; Faddoul et al. 2023 audit) — opacity ≠ neutrality
YouTube	…	…	…	Watch-time-maximising recommendation; rabbit-hole asymmetry documented for political content (Ribeiro et al. 2020, FAccT; Hosseinmardi et al. 2024 PNAS Nexus)
Telegram	…	…	…	Channel-broadcast model — no recommender, but high-velocity in-group amplification with no friction; documented as primary vector for narrative-laundering (NATO StratCom COE 2023)
Instagram / Threads / Reddit	…	…	…	Each has a documented engagement-curve asymmetry — fill in with platform-specific transparency-report citation

Cite the academic / transparency-report source on every algorithmic claim. "Platform X favours left/right" without citation is rejected at Pass 2.

🌍 Comparative-International Frame Lineage

No frame in Swedish politics is born locally. Trace the cognate frame to its prior appearance in another jurisdiction. Use public academic / EUvsDisinfo / Reuters Institute Digital News Report / GLOBSEC trend reports. Minimum 2 cognates per major frame; "no international cognate" for a significant story triggers re-do. Include regional reader notes for Asia, Americas, and Europe after the table.

Cognate frame	First major appearance	Vehicle	Mutation in Swedish context	Reference
`e.g., "great replacement" / "open-borders elites"`	France 2011 (Camus); US 2017 (alt-right)	Tabloids → talk-radio → social	Translated to Swedish "befolkningsutbyte" 2017–2019	EUvsDisinfo dossier; Reuters Institute 2024
`e.g., "elite vs. people" populist binary`	Hungary 2010 (Fidesz); Italy 2018 (Lega/M5S)	State media → friendly tabloids	Adapted by SD ahead of 2018, intensified 2022	GLOBSEC 2024 trend report
`e.g., firehose-of-falsehood doctrine`	RU 2014 (Crimea / MH-17)	RT/Sputnik network	Sweden 2022 NATO-application disinfo wave	RAND PE-198; SÄPO 2023 annual report
`e.g., "courts vs. democracy" frame`	Poland 2015 (PiS); Israel 2023 judicial reform	Government-aligned press	`[applies to current Swedish story? Y/N]`	Freedom House Nations in Transit 2025

Naivety check: if every frame in the table has "no international cognate found" — re-do the search. ≥ 2 cognates is the realistic floor for any major Swedish political story in 2026.

🎭 Strategic-Doctrine Detection

Identify whether the observed frame ecology fits a known propaganda doctrine. Detection ≠ attribution. Pattern-match to public-domain doctrines.

Doctrine	Signature	Observed?	Evidence
Firehose of falsehood (RAND PE-198, RU model)	High-volume + multi-channel + rapid + repetitive + no commitment to truth	🟢 / 🟧 / 🟥	…
Doppelganger operation (EU-DisinfoLab 2022 RU campaign cloning EU outlets)	Spoofed look-alike domains of named outlets	…	…
Gish gallop (debate technique scaled to media)	Many low-quality claims faster than fact-checking can rebut	…	…
Reflexive control (Soviet/RU doctrine)	Shape adversary's decision frame so they "freely" choose what attacker wants	…	…
Active measures spillover (NATO StratCom COE, 2024)	Domestic actors knowingly or unknowingly carrying foreign-state framing	…	…
Narrative capture by interest-group lobby (US tobacco / fossil-fuel model)	Coordinated industry-funded "expert" amplification	…	…
MAGA cognate populism (US 2016/2020/2024)	Anti-elite + media-as-enemy + perpetual-grievance + leader-cult components	…	…

No doctrine match is a valid finding. "Frames appear to be organic competitive politics within Swedish norms" is a publishable conclusion at HIGH confidence if evidence supports it.

⏳ Frame Lifecycle / Longevity

xychart-beta
    title "Frame reach over time — peak / half-life / decay"
    x-axis ["T-7d", "T-3d", "Today", "T+3d", "T+7d", "T+14d", "T+30d", "T+90d"]
    y-axis "Reach (normalised, 0–10)" 0 --> 10
    line [1, 4, 9, 8, 6, 4, 2, 1]
    line [0, 1, 3, 5, 7, 6, 4, 2]
    line [2, 2, 2, 3, 3, 4, 5, 7]

Line 1 = Frame A (peaked, decaying). Line 2 = Frame B (rising, lagged). Line 3 = Frame C (slow burn — gains over horizon).

Frame	Phase	Estimated peak (UTC)	Half-life (days)	Sleeper / zombie probability	Reactivation trigger
🟢 A	post-peak	`YYYY-MM-DD`	`e.g., 6`	LOW (15 %)	New SCB distributional data
🔴 B	rising	`YYYY-MM-DD`	`e.g., 9`	MED (35 %)	Court ruling on appeal
🟡 C	slow burn	`T+30d`	`e.g., 21`	HIGH (60 %)	EU Commission letter
🟣 D	episodic	…	`e.g., 3`	MED (40 %)	Coalition vote
⚫ E	dormant	n/a	n/a	LOW (10 %)	External crisis

Frame archaeology: for any frame with zombie probability ≥ MED, list the prior cycles where the same frame appeared (date + carrier). Zombie frames are the most predictable election-cycle weapons.

📈 Impact Conversion — RRPA (Reach × Resonance × Persistence × Action)

Frame	Reach (est. impressions, public metrics)	Resonance (engagement / impression)	Persistence (days at >50 % peak)	Action conversion (poll move / petition / protest / donation)	RRPA composite (0–100)
🟢 A	…	…	…	`e.g., +1.2 pp in M support, Demoskop (B2)`	…
🔴 B	…	…	…	…	…
🟡 C	…	…	…	`e.g., 2 EU MEP letters`	…
🟣 D	…	…	…	…	…

Action conversion is the only honest measure of frame power. Reach without action is theatre. Cite a dated indicator (poll, petition signature count, donation report, demonstration permit) for every Action cell, or write [no action signal yet].

🛡️ Counter-Resilience Plan — prebunking · inoculation · debunking ladder

Riksdagsmonitor itself does not push counter-frames — its only neutrality is procedural (equal analytical depth across all 8 Riksdag parties; equal evidentiary discipline across all outlets; full ownership/funding/lean disclosure for every cited outlet). This is not a claim that the platform produces "neutral truth" or that any media source is neutral — both claims are explicitly rejected by the founding doctrine at the top of this template. This section gives readers (journalists, policymakers, civic educators) a documented ladder so they can operate themselves.

Layer	Audience	Tactic	Reference
L1 — Prebunking (proactive)	General public, civic educators	Publish technique-focused frame-anatomy explainer 7+ days before predicted peak; show the manipulation technique (not a counter-political-position) (Roozenbeek & van der Linden 2022, Sci Adv)	https://prebunkingmovement.com/ ; Cambridge Social Decision-Making Lab
L2 — Inoculation (just-in-time)	Newsroom editors, fact-checkers	Distribute frame-fingerprint card before press conferences	UNESCO 2023 Journalism, fake news & disinformation handbook
L3 — Lateral-reading prompt	Readers	Provide outlet-ownership table (above) so readers can do source-triangulation themselves	Wineburg & McGrew (2019) Stanford HEG
L4 — Debunking (post-spread)	Journalists	Truth-sandwich format: truth → falsehood (briefly) → truth again	Lewandowsky et al. (2020) Debunking Handbook 2020
L5 — Algorithmic-friction request	Platforms	If state-affiliated coordinated amplification observed → file public DSA Art. 40 transparency request and EUvsDisinfo report	EU DSA Art. 40; Code of Practice on Disinformation 2022

Frame	Recommended counter-resilience layer(s)	Rationale
🟢 A	L3 + L4	Government frame is loud and durable — readers benefit from lateral-reading + truth-sandwich
🔴 B	L1 + L4	Opposition frame uses identifiable bias (in-group); prebunking the technique works
🟡 C	L2 only	Analytical frame doesn't require counter-resilience; inoculation guards against capture
🟣 D	L4	Coalition rumour responds to evidence-based debunking
⚫ E	L1 + L5	Foreign-overlay frames are textbook prebunking targets + DSA escalation

🔍 Quote Salience

Quote	Speaker	Frame	Reach (est.)	Reusability	Manipulation flag
`"verbatim, public source"`	`Named MP / outlet (party + role)`	A/B/C/D/E	High/Medium/Low	High/Medium/Low	None / `[doctored]` / `[out-of-context]`

🎯 Frame-Competition Dynamics

flowchart LR
    A["🟢 Frame A"] <-->|"contest"| B["🔴 Frame B"]
    C["🟡 Frame C"] -->|"adds risk"| A
    C -->|"strengthens"| B
    D["🟣 Frame D"] -->|"erodes"| A
    E["⚫ Frame E"] -.amplifies.-> B

    style A fill:#4CAF50,color:#FFFFFF
    style B fill:#D32F2F,color:#FFFFFF
    style C fill:#FFC107,color:#000000
    style D fill:#7B1FA2,color:#FFFFFF
    style E fill:#212121,color:#FFFFFF

Dynamic	Outcome	WEP confidence
Frame A vs B	`e.g., Stalemate — voter segments cluster`	LIKELY
Frame C interacts with both	`Adds structural risk to A`	LIKELY
Frame D adds noise to A	`Lowers government's narrative control`	EVEN
Frame E amplifies B	`IF observed — otherwise omit`	UNLIKELY (default)

📈 Coverage-Volume Dashboard

Outlet category	Day 1	Day 2	Day 3	Trend	Note
National daily press (SE)	11	14	9	↘	Transitioning to commentary phase
Tabloids (SE)	7	10	8	↘	…
Public broadcasters (SE)	4	5	3	↘	…
Commercial broadcasters (SE)	3	4	4	→	…
Regional press (SE)	22	18	12	↘	…
Opinion / commentary (SE)	6	9	11	↗	Frame C rising as economists weigh in
International quality press	1	2	4	↗	Reuters / Politico EU pickup
State-affiliated foreign	0	1	2	↗	⚠️ flag if non-zero — investigate amplification

🔁 Forward Watchlist

Trigger	Likely frame shift	WEP	Time horizon	Admiralty
EU Commission formal communication	Frame C surges to dominance	LIKELY	14 d	A1 (forthcoming primary)
SCB publishes distributional analysis	Frame B strengthens	HIGHLY LIKELY	30 d	A1
Government announces follow-on measure	Frame A strengthens	EVEN	30 d	B2
Coalition-party dissent visible	Frame D surges	UNLIKELY	7 d	C3
EUvsDisinfo dossier on this story	Frame E confirmed if observed	UNLIKELY (default)	30 d	A1
SÄPO public statement	Frame E + foreign-actor attribution	VERY UNLIKELY	90 d	A1

🎯 Personalization & Character-Attack Analysis (target = the ATTACK, never the private person)

ETHICAL FIREWALL — non-negotiable, Pass-2 blocking. This section analyzes the mechanics of a personal attack as a media-manipulation phenomenon. It does NOT profile, investigate, or repeat allegations about private individuals.

Public officials are analyzable only in their official capacity (GDPR Art. 9(2)(e)/(g)).

Private individuals — spouses, partners, children, named private citizens — are NEVER the analytical object. They appear only as the target of a documented attack, and only to expose the manipulation.

Any unverified personal allegation is a manipulation datapoint to be flagged — NEVER a claim to be restated as fact. Repeat nothing; map the attack vector.

Personalization Vector Table

Attack vector	Target class	Legitimate scrutiny OR smear?	Evidence (public URL / dated)	Manipulation technique	Private-individual shield applied?
Record/conduct in office	Public official (official capacity)	Legitimate if evidenced	`[url]`	— (fair scrutiny)	N/A
Character / integrity smear	Public official	Smear if unevidenced	`[url]`	`T0023 Distort facts`	N/A
Family-member instrumentalization	Private individual (spouse/relative)	Smear — always	`[url — the ATTACK, not the allegation]`	Guilt-by-association	✅ shielded — not profiled
Gendered / sexualized attack	Public official + private	Smear	`[url]`	`T0048 Harass`	✅ if private target
Deepfake / synthetic media	Either	Smear	`[url]`	`T0088 AI-generated media`	✅ if private target

Attack-Escalation Ladder (severity classification)

flowchart LR
    S1["🟢 L1 Legitimate scrutiny<br/>evidenced, official-capacity"] --> S2["🟡 L2 Character framing<br/>unevidenced integrity attack"]
    S2 --> S3["🟠 L3 Guilt-by-association<br/>family/relationship weaponized"]
    S3 --> S4["🔴 L4 Harassment / smear campaign<br/>coordinated, CIB signals"]
    S4 --> S5["⚫ L5 Incitement / threat<br/>escalate to reporting note"]
    style S1 fill:#4CAF50,color:#FFFFFF
    style S2 fill:#FFC107,color:#000000
    style S3 fill:#FB8C00,color:#FFFFFF
    style S4 fill:#D32F2F,color:#FFFFFF
    style S5 fill:#212121,color:#FFFFFF

Public media coverage and public social-media posts only. Representative sample across Swedish national, regional, and commentary outlets, plus international quality press for comparative frame lineage. No outlet is treated as neutral — every cited outlet appears in the §"Source Ecology — Outlet Bias Audit" table with ownership, funding mix, board-appointment authority, documented editorial lean, Reuters Institute Trust score, and PO/PON complaint history. Public-service broadcasters (SVT/SR/UR) are publicly funded with politically-appointed boards (Förvaltningsstiftelsen) and an institutional editorial culture — process-disciplined but not bias-free. State-affiliated outlets (RT/Sputnik/CGTN/PressTV/RIA/TASS) monitored only as amplification signal — never cited as factual sources. No paywall bypass. No private-account social-media content. No leaked or hacked material. Frame-ownership and funding data sourced from Nordicom Media Ownership Database (gu.se), Reuters Institute Digital News Report 2024, EUvsDisinfo case dossiers, EU DSA transparency reports, and public SÄPO/MUST/FRA/EU-EEAS publications.

Document Control

Template path: /analysis/templates/media-framing-analysis.md
Version: 2.3 (preamble notices removed; production-mode info moved to Framing Context; version badge corrected — 2026-05-03)
Previous version: 2.2 (global audience orientation; multi-dimensional alignment framework — 2026-05-02)
Referenced by: ai-driven-analysis-guide.md § Step 5, per-artifact-methodologies.md § media-framing-analysis, electoral-domain-methodology.md § Part 5
Classification: Public
Next Review: 2026-08-01

✅ Pass-2 Self-Audit Checklist (v2.3 — required)

Purpose: AI-FIRST principle requires a Pass-2 read-back-and-improve. After producing this artifact in Pass 1, re-read it end-to-end and verify each item below. Document any remediation in methodology-reflection.md §"Pass-2 audit log". Any unchecked ❌ box at the end of Pass 2 forces a Pass-3 rewrite of the affected section.

Global Audience & Multi-Dimensional Alignment (NEW in v2.2 — top priority)

Global Audience Orientation section present and populated — Swedish political-system context (Riksdag 349 seats, 4 % threshold, 8 parties, coalition, election date + days-remaining) provided for readers with no prior knowledge of Swedish politics; regional reader notes for Asia, Americas, and Europe present.
Multi-dimensional alignment used throughout — outlet editorial lean in the Outlet Bias Audit uses the 5-axis framework (economic / social-identity / EU / security / media-ownership), not a single left/right score.
Comparative-international frame lineage includes global notes — at least one Asia note, one Americas note, and one Europe note per major frame or as a block after the lineage table.
No single left/right score in the artifact body — any sentence characterising an outlet as simply "left" or "right" without axis qualification is rewritten at Pass 2.

No-Neutral-Media Doctrine (v2.1 — non-negotiable)

No "neutral / impartial / unbiased / balanced / objective" label applied to any outlet, algorithm, or broadcaster in the artifact body without ≥ 1 documented-bias citation in the same paragraph. Free-text occurrences of these words in any analytical sentence trigger Pass-2 rewrite.
Outlet Bias Audit table is populated for every cited outlet — ownership group, funding mix, board-appointment authority, multi-dimensional editorial lean, Reuters Institute Trust score, PO/PON complaint history, foreign-actor link. No bias audit row = no citation.
Public-service broadcaster caveat present — every reliance on SVT / SR / UR / BBC / ARD / NRK / DR / YLE is accompanied by the licence-fee + politically-appointed-board + institutional-editorial-culture caveat, not treated as ground truth.
Algorithmic-asymmetry rows declare optimisation target — every platform row names the objective function (engagement / watch-time / retention / ad-yield) and cites a documented partisan/emotional asymmetry; "the algorithm" without optimisation-target attribution is rejected.
Frame C / D label discipline — Frame C is labelled "establishment / centrist-consensus" (NOT "neutral / analytical / institutional"); Frame D is labelled "public-broadcaster proceduralist" with politically-appointed-board + institutional-editorial-culture caveat (NOT "neutral / impartial / objective reporting" / "what really happened" / "coalition-inside").
Procedural neutrality only — any reference to platform "neutrality" is qualified as procedural (equal analytical depth across all 8 Riksdag parties; equal evidentiary discipline across all outlets; full ownership/funding/lean disclosure) and never as a claim of "neutral truth" or "neutral source".