Compare governments
How much of each government's agenda goes to a subject, next to the regional average. Counting alone would rank these places by size — Cary puts ten times as many items on its agendas as Pittsboro — so everything here is a share or a rate, and the count it rests on is always beside it.
Search and filter
Of the agenda items we have been able to classify for this government, what fraction carry the subject. Not of every item — whether an item is classifiable at all depends on the agenda platform, not on the government.
Litigation and settlements, past 12 months
| Government | Share of classified work | Items | Of classified | Agenda read | Range | Against the region |
|---|---|---|---|---|---|---|
| Chatham County | 2.1% | 10 | 462 | 44% | 1–4% | more than the region |
Download this table as CSV Two governments side by side What has been changing
How to read this
Share is the fraction of a government's classified agenda items carrying the subject — not of every item. That distinction matters more than it sounds. Where a portal publishes a structured item list, that list is the whole agenda, call to order and adjournment included; where it does not, the item list is written out by the summarization worker, which records the substantive business and skips the ceremony. On this site 40% of Cary's items carry a subject against 79% of Apex's, and the difference is the agenda platform, not the town. Dividing by every item made portal governments look half as engaged in everything.
The share is adjusted toward the regional average in proportion to how little evidence stands behind it — a board with four classified items and one rezoning does not get reported as the most development-focused place in the region. The range beside it is a 95% interval on the unadjusted figure.
Against the region says whether that interval clears the regional average. Agenda items are not a random sample of anything, so this is a guard against reading noise as a finding rather than a statistical test. A government with few items will almost always read as within the usual range, because with few items almost nothing can be told apart.
Per 10,000 voters is judged against a per-resident regional rate, not against the share — a small town with a busy agenda can be first per resident and unremarkable as a share, and both are true. The interval there is on the item count rather than a percentage, because once the denominator is a population the items are arrivals and not a proportion. It uses registered voters from the State Board of Elections, summed across counties for a town that straddles a line. Registered voters are not population: they leave out children, non-citizens, and adults who have not registered, and the share who register differs from place to place. It is the denominator this state publishes per municipality and keeps current, which no population estimate does.