Indonesia: election results by polling station

2019 · 1 election · IDN

SAGE has legislative and presidential election results for Indonesia from 2019. The smallest unit is the polling station. The 2019 election has 806,079 polling stations, and 88% of them have a location. Unit boundaries are available for every election.

[explore on the map] [download] [codebook] [all 134 countries]

Elections

Units are the smallest units in the data, and located units are those with coordinates. A year with several rows has several contests or votes; filter on them before you add up votes.

Year	Election type	Vote type	Round	Units	% located	Votes	Polygons
2019	Legislative (lower house)			799,266	88%	136,686,753	yes	[map]
	Presidential			806,079	88%	152,171,781	yes

Download

csv opens in Excel, Stata, R, or Python. gpkg has the unit boundaries for QGIS, ArcGIS, R, or Python. parquet has the vote rows as Parquet. rds is the native R sf object. Each ZIP has a README with the columns.

All years: [csv 154 MB] [gpkg 145 MB] [parquet 213 MB] [rds 53 MB] [codebook 1.6 MB]

Year	Downloads
2019	[csv 154 MB] [gpkg 145 MB] [parquet 64 MB]

In R or Python:

R:      install.packages("sage")
        sage::sage_load("Indonesia", years = 2019)

Python: import sage
        sage.sage_load("Indonesia", years=[2019])

Sources

Komisi Pemilihan Umum Republik Indonesia

Notes

Villages are matched to boundaries via a two-step process, whereby the name of the village is geocoded and the resulting latitude-longitude are matched to village boundaries.
Identifier code NAME columns (i.e., NAME1_b, NAME2_b, etc.) change between 2019 and 2024 election, but the names of places themselves (provinsi, kota, kecamatan, kelurahan) are largely consistent across years.
Supplementary shapefiles for Indonesia are provided at the NAME4 (village) level.

Citation

Dasanaike, Noah. “The Small-Area Global Elections (SAGE) Dataset.” Scientific Data (2026). Paper · Harvard Dataverse

← India · All countries · Iran →