Parliament activity

See who does what in the Eduskunta: plenary speeches, written and oral questions, interpellations, and motions. Toggle between speeches and questions & initiatives, group by party or by MP, and normalize by seats or sitting MPs to compare fairly across party sizes. Covers 2019–2026 across the Marin and Orpo terms.

Speech figures describe the selected analytical sample, not every parliamentary speech. See dataset scope.

Speeches come from plenary puheenvuoro records; the type buckets collapse the Finnish labels into six groups (Regular, Reply, Fast, Group, Intro, Other). Questions & initiatives are first-signer matters: KK written, SKT oral, VK interpellation, TPA/TAA/KA motions. Seat counts are from the official Vaalit.fi results (2019, 2023). "Sitting MPs" counts MPs who gave at least one speech in the term. Note the systematic government/opposition effect: governing parties file far fewer questions because they don't interrogate themselves. In the all-years view, per-seat figures divide by the combined seats across both electoral terms (seat-terms), while per-MP figures use distinct speakers across the whole period.

Which party talks about what?

Each party's share of speeches per area, expressed as an over-index (party share ÷ area's share across the whole chamber). A 3.6× cell means: this party talks about this topic 3.6× more often than the chamber average. Dataset: 41,528 area-tagged speeches.

Over-index = party's speech share on an area ÷ the area's share across the whole chamber for the same term. The baseline is fixed to the whole chamber so the government/opposition filter stays on the same scale. Figures use the published area dataset (full sample).

Ideological compass

Each party is plotted from 20,763 accepted axis scores from 18,319 distinct speeches. X-axis: economic (−2 left … +2 right). Y-axis: GAL-TAN (−2 cosmopolitan/green … +2 traditional/nationalist). Bubble size ∝ number of accepted axis scores.

Every sufficiently long speech is evaluated separately on the economic and GAL-TAN axes. The model may mark either or both axes unscorable when the speech contains no explicit position from the speaker. Accepted −2…+2 axis scores are averaged by party. Speeches were scored by deepseek-v4-flash; the party is withheld from the prompt. The prompt is in the about section. References are scaled to the same −2…+2 axis: CHES (0–10 GAL-TAN → −2..+2 via (x-5)·0.4); Manifesto (RILE -100..+100 → -2..+2 via x/50); EVA (Economic/social -1..+1 → -2..+2 via ×2); Ohjelmat (−2..+2 per-party mean across all programmes).

Reference datasets

Parliamentary speeches (LLM)

Speech-derived party coordinates (plenary speeches 2019–2026). (−2..+2 per-party mean of 20763 axis scores (18319 speeches), model deepseek-v4-flash)

CHES

Chapel Hill Expert Survey: Expert judgments on party positions (2015 trend file). (0–10 GAL-TAN → −2..+2 via (x-5)·0.4)

Manifesto

Manifesto Project (MARPOR): Automated coding of party election manifestos (RILE index). (RILE -100..+100 → -2..+2 via x/50)

EVA

EVA value survey: Finnish citizen value survey ('Suomen kuvitettu arvokartta'). (Economic/social -1..+1 → -2..+2 via ×2)

Ohjelmat

Party programmes (LLM): LLM-scored party principle & election programmes (Pohtiva archive), same axes. (−2..+2 per-party mean across all programmes)

How do position and agenda shift over time?

Two signals per parliamentary year (2019–2026): each party's mean position on each ideological axis, and the whole chamber's agenda shares by area. Built from 20,763 accepted axis scores from 18,319 distinct speeches.

Election answers and parliamentary work

Pre-election answers from the 186 sitting MPs who answered Yle's eduskuntavaalit 2023 vaalikone, projected onto the economic and GAL-TAN axes. These are the same axes used by the compass. The answers are then compared with where their Orpo-term parliamentary work lands. The economic and GAL-TAN classification is done by a large language model. The arrow shows drift between the election-compass and parliamentary-work positions; the salience bars show whether their floor focus matched what they felt strongest about before the election.

This comparison shows alignment signals derived from the available data and model-assisted mappings; it is not a definitive assessment of promise fulfilment.

Promised coordinates are projected from each MP's 30 vaalikone answers: every question is hand-mapped to one axis (economic or GAL-TAN) and a sign (whether "agree" points left/GAL or right/TAN). Each answer's (value − 3) is multiplied by the sign and averaged across that axis's questions, yielding a −2…+2 coordinate directly comparable to the speech stances. Speech-position coordinates are the mean of each MP's per-speech stance scores during the Orpo term (2023–). Drift is the euclidean distance between the two points. The headline drift measures broad speech alignment. The question-level audit keeps topic matches separate from compliance verdicts. A vote match or contradiction requires a mapped vote: the mapping is LLM-suggested and direction-validated against the party Jaa/Ei split. A same-direction marker is the weakest signal: the MP spoke in the area in a consistent direction, without addressing this specific claim.

About

• Speeches
  41,528 plenary speeches, api.eduskunta.fi puheenvuoro search. ~50% sample (per-MP cap of 200/year).
• MPs
  281 distinct speakers with recorded speeches.
• Questions & initiatives
  8,266 party-attributed matters from parliamentary years 2019, 2020, 2021, 2022, 2023, 2024, 2025, 2026: written and oral questions (KK, SKT), interpellations (VK), and motions (TPA, TAA, KA).
• Stance scores
  20,763 accepted axis scores from 18,319 distinct speeches. Every speech is evaluated separately on the economic and GAL-TAN axes.
• Area tags
  Keyword mapper (full coverage); a language-model classifier validates a subset.
• Parliamentary years
  2019, 2020, 2021, 2022, 2023, 2024, 2025, 2026

Area classification

Every speech and matter is assigned one policy area from a fixed taxonomy of 13 categories. The default is a deterministic Finnish keyword mapper over the matter title. It is free and instant, with about 75% coverage. A language-model classifier tags a subset for higher per-speech precision; the party is withheld from the prompt.

Stance scoring

Every sufficiently long speech is evaluated separately on both axes using a scale from −2 to +2. If a speech contains no explicit position on an axis, that axis is marked unscorable and excluded from the mean:

The prompt instructs the model to reason from speech substance, not speaker identity, and marks speeches without an explicit position as unscorable. Party is withheld to prevent speaker-prior bias. The published scores used deepseek-v4-flash. The prompt is available below.

Olet suomalaisen politiikan asenneanalyytikko. Sinulle annetaan yksi
eduskuntapuheenvuoro. Arvioi puhujan oma kanta ERIKSEEN kahdella akselilla.

TALOUSAKSELI (economic), vasen ↔ oikea:
−2 = vaatii julkisen sektorin, uudelleenjaon tai työntekijän aseman LAAJENTAMISTA
     (lisää palveluita, korkeampi verotus, vahvempi sääntely/ay-asema)
−1 = puolustaa nykyisiä julkisia palveluita tai etuja leikkauksia vastaan
 0 = nimenomaisesti keskitie tai tasapaino talouskysymyksessä
+1 = kannattaa maltillista markkinaehtoisuutta, veronkevennystä tai menokuria
+2 = vaatii selvästi markkinaehtoisuutta, merkittäviä veronkevennyksiä tai
     julkisen sektorin supistamista

ARVOAKSELI (gal_tan), GAL ↔ TAN:
  GAL = vihreä, vaihtoehtoinen, liberaali, kosmopoliittinen
  TAN = perinteinen, autoritaarinen, nationalistinen
−2 = vahva GAL (yksilönvapaus, avoimuus, pluralismi, yli-rajainen yhteistyö)
−1 = maltillinen GAL
 0 = nimenomaisesti keskitie tai tasapaino arvokysymyksessä
+1 = maltillinen TAN
+2 = vahva TAN (kansallinen sulkeutuneisuus, järjestys, perinteiset normit)

Säännöt jokaiselle akselille:
- Perusta arvio VAIN puhujan omaan, eksplisiittiseen kantaan. Älä käytä puoluetta,
  henkilön tunnettuja mielipiteitä tai yksittäisiä avainsanoja oikoteinä.
- Hallituksen puolustaminen tai opposition arvostelu EI OLE oma kanta. Kysy:
  esittääkö puhuja oman vaihtoehdon, vai reagoiko hän vain toisten esityksiin?
  Jos jälkimmäinen, akseli on "unscorable".
- Useimmat puheet ottavat kantaa vain TOISEEN akseliin. Se on odotettua: merkitse
  toinen "unscorable". Älä pakota kantaa akselille, jota puhe ei käsittele.
- Aihe ei ratkaise akselia: talouspuhe voi ottaa arvokannan ja päinvastoin.

status jokaiselle akselille:
- "clear": selvä oma suunta; stance −2, −1, +1 tai +2.
- "neutral": puhuja kannattaa eksplisiittisesti keskitietä; stance 0.
- "mixed": vastakkaisia omia kantoja ilman hallitsevaa suuntaa; stance 0.
- "unscorable": tähän akseliin liittyvää omaa kantaa ei ole; stance 0.

Vastaa kentät tässä järjestyksessä jokaiselle akselille: poimi ensin SANATARKKA
lainaus puheesta ("evidence"), päätä sitten status ja stance. Älä tiivistä tai
muotoile lainausta uudelleen; jos sanatarkkaa lainausta ei ole (esim. unscorable),
jätä "evidence" tyhjäksi.

AIHE (taustatieto): {matter}

PUHEENVUORO:
{speech}

Party compass

A party's place on the compass is the mean of its stance-scored speeches on each axis. Each party therefore gets one point: (economic, GAL-TAN). The same coordinates are compared below to CHES, Manifesto, and EVA.

Activity

The activity view counts plenary speeches as well as questions and initiatives (by first signer). Figures can be shown by party or by MP, as raw counts or relative to seat totals or sitting MPs — otherwise larger parties win on size alone.

Seat counts come from Vaalit.fi (2019 and 2023). Here, the Marin term means parliamentary years 2019–2022 and the Orpo term 2023–2026. The all-years per-seat figure uses the combined seat count across both terms.

Our speech-derived compass agrees strongly with three independent reference datasets (Pearson r, 8 shared parties). Each uses a fundamentally different methodology:

1. Eduskunta open-data API: api.eduskunta.fi. Valtiopaivaasia and Puheenvuoro categories.
2. CHES: Chapel Hill Expert Survey. Expert judgments on party positions (2015 trend file). (0–10 GAL-TAN → −2..+2 via (x-5)·0.4)
3. Manifesto: Manifesto Project (MARPOR). Automated coding of party election manifestos (RILE index). (RILE -100..+100 → -2..+2 via x/50)
4. EVA: EVA value survey. Finnish citizen value survey ('Suomen kuvitettu arvokartta'). (Economic/social -1..+1 → -2..+2 via ×2)
5. Ohjelmat: Party programmes (LLM). LLM-scored party principle & election programmes (Pohtiva archive), same axes. (−2..+2 per-party mean across all programmes)
6. YSO, the Finnish General Ontology: finto.fi/yso.