# Postal Codes Dataset for Philippines (PH)

This dataset contains Philippine ZIP codes — 2,317 rows covering 2,190 distinct postal
codes across 87 provinces and districts in all 17 administrative regions.

**Granularity note:** ZIP is not one-to-one with place. Metro Manila codes in particular
are shared across several named barangays and post-office stations, so one code
legitimately appears on many rows. The primary key is (`postal_code`, `place_name`),
never `postal_code` alone.

**Institutional codes:** 280 rows carry no region, province or city. The Philippines
assigns ZIP codes to large institutional mailers and PO Box holders as well as to places;
on those rows `place_name` is an organisation and the source supplies no administrative
assignment. Those cells are blank rather than guessed. Filter on a populated
`admin_name1` if you need geographic rows only.

**Coordinate accuracy:** 644 rows have `accuracy` 1, meaning the point was interpolated
by the source from neighbouring postal codes rather than matched to a gazetteer entry,
and can fall outside the province the row names. Check `accuracy` before relying on a
coordinate.

**Note:** This dataset is a sample containing the first 100 rows. If you need access to the complete dataset, please contact our sales team for more information: datahub@datopian.com

## Data Sources

Postal codes, place names and coordinates are sourced from GeoNames (CC BY 4.0), a
community-maintained open geographic database. `admin_code1` is the official
ISO 3166-2:PH region code with the `PH-` prefix stripped, applied consistently to all 17
regions including the National Capital Region (`00`). Note there is no region `04` — the
old Region IV was split into `40` (CALABARZON) and `41` (MIMAROPA) — and `14` is BARMM
while `15` is the Cordillera Administrative Region, not the reverse.

## Usage

The dataset files are available in CSV format and can be accessed and downloaded using the links provided in the `datapackage.yml`.

`postal_code` and `admin_code1` carry significant leading zeros and must be read as text
(`dtype=str`), or use the schema in the packaged `datapackage.json`.

For access to the full dataset or other inquiries, please contact our sales team at datahub@datopian.com