PolitIQ Kenya · Nairobi · 2026-08-19
In short. Pressure on a Kenyan officeholder is not the volume of talk about them. It is hostility with weight behind it, in places they answer for, rising. We score it out of 100 from three factors, weighted 45 / 35 / 20.
The obvious instrument — count the mentions — fails in a way that is easy to miss. A president is discussed everywhere, constantly, most of it neutral. Score on volume and the capital city always tops the map, whatever is happening in it.
We found this in our own build. An early version of the index summed its three components, and returned a pressure score of 30 out of 100 for a county in which the officeholder had attracted no negative commentary at all. That is coverage wearing the label of pressure. Volume now multiplies hostility rather than adding to it: it can amplify a signal, never manufacture one.
| Factor | Weight | What it captures |
|---|---|---|
| What people say | 45% | Share of discourse about the officeholder that is hostile, damped by a prior so that a handful of angry comments cannot spike the index, then scaled by how much discussion sits behind that judgement. |
| What happens on the ground | 35% | Protests, strikes, assemblies, security deployments and communal conflict, weighted by severity. |
| What it costs to live | 20% | Cost-of-living conditions and economic grievance. In Kenya this drives mobilisation more reliably than partisan dispute does. |
A governor is under pressure when their county has marches and the price of unga is climbing, whether or not anyone writes their name. A national officeholder answers for all 47 counties. Someone out of office answers for the street far less — unrest is not a burden on the opposition, it is an opportunity — so ground and cost factors are attenuated for those who do not govern.
This is why two national figures do not produce the same map: the country's conditions are shared, the responsibility for them is not.
Below a minimum sample, a county returns nothing rather than zero. The distinction carries the weight of the whole product. Zero asserts that there is no pressure in a place. Absence says we cannot see it — usually because collection there is thin, which is the opposite conclusion for anyone deciding where to go next.
Our maps therefore report how many counties fell below sample alongside the counties that scored. A map with few marks is a statement about our coverage, and we would rather say so than let it read as a quiet country.
The standards this reporting is written to are set out in our analytic standards.