The numbers behind the mosque directory

Four countries, side by side

OpenStreetMap data via the Niyat mosque directory, 23 August 2026; counts reflect mapped mosques, not all mosques. Percentages are of each country's listed mosques; the nearest-neighbour figure is the median distance to the closest other mapped mosque in the same country.

Methodology

Source. All figures are computed from data/mosques.json, the OpenStreetMap-derived extract behind the Niyat mosque directory (23 August 2026 build), and the directory index of published pages. OpenStreetMap coverage is partial: many mosques are not yet mapped, mapped mosques may lack tags, and tags can contain errors. Nothing here is a census of mosques; counts reflect mapped mosques, not all mosques.

Which mosques are counted. A directory page exists when a mapped record carries both a name and a city (addr:city). Records missing either are excluded and reported per country. City totals group the addr:city value the same way the directory builds its city pages.

States and regions. United States, Canadian, and Australian tables use the OpenStreetMap addr:state / addr:province tags where present. United Kingdom regions are derived from the postcode area of addr:postcode (for example BD maps to Bradford, in Yorkshire and the Humber) using the standard twelve nations and regions. Coverage of these fields is uneven; each table states how many records it covers.

Facilities and contact. Shares are computed over the country's listed mosques. Website and phone reflect the presence of those tags; wheelchair shows the tag's values (yes, limited, designated, no); women's facilities reflect the small number of female / women tags. An untagged mosque may still have the facility — absence of a tag is not absence of a feature.

Denomination. Shown only where the denomination tag exists; the share of mosques with no tag is listed alongside.

Nearest-neighbour distance. For every mapped mosque, the great-circle (haversine) distance to the closest other mapped mosque in the same country, found with an approximate spatial-grid search in plain Python (no external libraries). The median is taken over mosques with at least one mapped neighbour.

Licence and attribution. The underlying map data is © OpenStreetMap contributors and available under the Open Database Licence (ODbL). If you republish figures, keep the caveat that counts reflect mapped mosques, not all mosques.

OpenStreetMap data via the Niyat mosque directory, 23 August 2026; counts reflect mapped mosques, not all mosques.

Cite this page

Linking to niyatapp.com/mosques/statistics/ is welcome; if you republish figures, keep the sentence that counts reflect mapped mosques, not all mosques.

Keep going