Our World in Data Population Density
- Rows
- 76,576
- Columns
- 4
- Data range
- -1000 to 900
- Auto-refresh
- Re-checks its source every 30 days
- Last refreshed with new data
- 2026-10-25
Population density time series by country or region, long-term historic data for correlations with lat/lon keyed datasets.
AI analysis
Dataset Analysis: Our World in Data — Population Density (Lat/Lon)
1. Dataset Overview & Value for Correlation Studies This dataset provides a long-term historical time series of population density by country or region, sourced directly from Our World in Data's public API (https://ourworldindata.org/grapher/population-density.csv). The publisher — Our World in Data — is a well-regarded academic-adjacent organization affiliated with Oxford University, known for rigorous sourcing and long-horizon data curation, which lends strong credibility to the figures. With 341 distinct years and 251 distinct entities across 76,576 rows, this dataset spans a remarkable temporal range, making it exceptionally well-suited for longitudinal correlation studies. Its explicit framing around Lat/Lon keyed datasets signals its intended purpose: serving as a joinable backbone for geographic correlation analyses where population pressure is a confounding or explanatory variable.
2. Data Quality Observations Data quality here is notably strong. There are zero null cells across all four columns, which is rare for a dataset of this size and temporal breadth — a direct reflection of Our World in Data's structured API output rather than scraped or manually compiled sources. Duplicate row counts are pending Phase D recomputation against Parquet, so that figure should be validated before treating the row count as fully authoritative for aggregations. No type mismatches are flagged, and all columns appear to have resolved cleanly to their intended types (Date, String, String, Decimal). The one quality caveat worth noting is that 251 distinct Code values perfectly matching 251 distinct Entity values confirms a clean 1:1 code-to-name mapping — no orphaned codes or unnamed entities are present.
3. Key Column Distributions — Population Density The Population density column is the analytical core and tells a vivid story through its statistics. The mean of 207.8 people/km² is dramatically inflated relative to the median of 28.5, a gap driven by extreme urban and small-island nation densities. The interquartile range (Q1=5.4, Q3=99.3) confirms that 50% of all observations fall below 100 people/km², meaning most country-year combinations represent sparsely populated territories. The maximum of ~22,797 people/km² (likely small high-density entities like Monaco, Singapore, or Macau) and a skewness of 12.6 — heavily right-tailed — underscores the need for log-transformation before any regression or correlation work. Most critically, 8,680 flagged outliers (~11.3% of rows) are not data errors but structural features of the distribution; analysts should segment small city-states and microstates from continental nations to avoid distorting correlations.
4. Recommended Join Key Columns The natural join strategy here is a composite key on Code + Year. Code appears to follow the standard ISO 3166-1 alpha-3 format (e.g., USA, GBR, CHN), which is the most widely used country identifier in global datasets — enabling clean joins without fuzzy string matching on Entity names (which vary across publishers). Year as a Date type should be validated for exact format (YYYY vs. YYYY-MM-DD) before joining, as mismatches in temporal granularity are a common join failure point. For purely spatial (non-temporal) correlations, Code alone as a single key with a chosen anchor year (e.g., 2020) is recommended.
5. Suggested Pairing Datasets for Correlation Discovery This dataset pairs naturally with several high-value correlation targets: - Climate & Geography: Temperature/precipitation grids or elevation datasets keyed by lat/lon centroid — directly aligned with the dataset's stated Lat/Lon purpose — to test whether population density correlates with temperate climates or coastal proximity. - Economic Development: GDP per capita or Human Development Index (HDI) by country-year to explore the density-prosperity relationship across time. - Health Outcomes: Disease prevalence, life expectancy, or pandemic spread rates (e.g., COVID-19 case density) where population density is a primary transmission covariate. - Land Use / Deforestation: Forest cover change or agricultural land percentage to examine how density drives land pressure. - Infrastructure & Urbanization: Access to electricity, internet penetration, or urbanization rates — all available from World Bank APIs with matching ISO country codes and annual granularity.
Columns
- Year (date)
- Code (string)
- Entity (string)
- Population density (decimal)