Sentiment Analysis
How Kenyans are reacting to Parliament across mainstream media. Limited to vetted sources — see our methodology.
Mood Over Time
Sentiment by topic — show the data table
A single item can mention several topics, so these mentions sum to more than the 35 items analysed.
| Topic | Mentions | Average sentiment |
|---|---|---|
| Devolution | 11 | -0.45 |
| Education | 11 | -0.19 |
| Governance & Constitution | 10 | 0.03 |
| Economy & Taxation | 10 | -0.40 |
| Security | 8 | -0.79 |
| Healthcare | 6 | -0.20 |
| Corruption & Accountability | 4 | -0.45 |
| Agriculture | 2 | -0.15 |
By Source
| Source | Items | Average sentiment |
|---|---|---|
| The Standard | 30 | -0.29 |
| Nation | 4 | -0.41 |
| The Star | 1 | 0.60 |
Recent Voices
How This Is Measured
We track how Kenyans react to Parliament using coverage from reputable mainstream media — established national newspapers and the official channels of established TV and radio newsrooms. Every few hours our pipeline ingests new articles and the public comments on those newsrooms' official YouTube channels.
Each item is matched to policy topics (such as Healthcare, Education and the Economy) and then classified as positive, neutral or negative — with a score from −1.0 to +1.0 — by a large language model that handles sarcasm and mixed sentiment conservatively. Scores are then averaged by topic, by source and over time, and everything above is filterable by date. Sentiment is indicative, not definitive. For the full process and the list of sources, see the methodology.