GINI Index (Historic, World)
- Rows
- 2,351
- Columns
- 4
- Data range
- 1963-01-01 to 2024-01-01
- Auto-refresh
- Re-checks its source every 30 days
Historic values of the GINI inequality index for countries worldwide, sourced from the World Bank. Measures income distribution inequality over time. ODC-PDDL license.
AI analysis
GINI Index (Historic, World) — Dataset Analysis
1. Dataset Overview & Research Value This dataset captures the GINI inequality index — a globally recognized measure of income distribution where 0 represents perfect equality and 100 represents perfect inequality — across 171 countries spanning six decades (1963–2024). Sourced via the World Bank API and hosted at [datahub.io/core/gini-index](https://datahub.io/core/gini-index/r/-/data/gini-index.csv), the publisher (DataHub.io / Frictionless Data) maintains this as a curated "core" dataset under the open ODC-PDDL license, suggesting reasonable freshness and institutional reliability. With 62 distinct years and 171 countries represented, this dataset has strong longitudinal and cross-sectional potential, making it particularly well-suited for studying how inequality evolves alongside macroeconomic shifts, policy changes, and development milestones.
2. Data Quality Observations The most significant quality concern is the extreme sparsity of the Value column, which carries 2,108 nulls out of 2,351 rows — a staggering 89.7% null rate. This reflects a well-known characteristic of GINI data: national income surveys are expensive and infrequent, so most country-year combinations simply have no recorded measurement. The structural columns (Year, Country Name, Country Code) are clean with zero nulls, which means the dataset's skeleton is intact, but analysts should treat this more like a sparse panel dataset than a complete time series. Duplicate row counts are noted as pending recomputation, but given the compound join key (Year + Country), duplicates are unlikely to be a major issue. The integer typing of Value is worth flagging — GINI scores are typically reported as decimals (e.g., 34.7), so some precision may have been lost during ingestion.
3. Key Column Distributions The Value column, despite its sparsity, tells a meaningful story across its 243 non-null observations. The mean of 37.23 and median of 35 suggest a modest right skew (skewness = 0.849), confirmed by the asymmetric interquartile range: Q1=30, Q3=43, with the upper tail pulling the mean above the median. The min of 23 and max of 63 bracket a wide real-world range — scores in the low 20s typically reflect highly egalitarian Nordic or Eastern European economies, while scores near 63 align with historically high-inequality nations in sub-Saharan Africa or Latin America. The standard deviation of ~8.97 reflects genuine global heterogeneity rather than noise. Only 3 outliers were flagged, suggesting the extremes are real data points rather than errors. The 37 distinct integer values across 243 observations also reinforces the precision-loss concern noted above.
4. Recommended Join Key Columns The optimal join strategy uses a compound key of Country Code (ISO 3-letter) + Year. Country Code is strongly preferred over Country Name for cross-dataset joins, as country names vary widely in spelling and formatting across sources (e.g., "Korea, Rep." vs. "South Korea"). The Year field, stored as a full ISO timestamp (e.g., 1963-01-01T00:00:00), will need to be truncated to the year integer before joining with most macroeconomic datasets. Given the sparse coverage, analysts should plan for left or outer joins rather than inner joins to avoid inadvertently discarding country-years where GINI data exists but partner dataset values do not.
5. Suggested Complementary Datasets for Correlation Analysis This dataset pairs exceptionally well with several thematic neighbors: - GDP per Capita (World Bank / IMF) — the classic Kuznets Curve hypothesis predicts an inverted-U relationship between income and inequality, directly testable here. - Human Development Index (UNDP) — correlating GINI with HDI sub-components (education, health, life expectancy) can reveal whether inequality undermines human development outcomes. - Tax Revenue as % of GDP — tests whether redistributive fiscal policy correlates with lower GINI scores across countries. - Social Protection Spending — welfare expenditure datasets (ILO, World Bank) could reveal policy levers that dampen inequality. - Political Regime / Democracy Index (V-Dem, Freedom House) — exploring whether democratic governance correlates with lower inequality over time is a well-studied but data-hungry research question that this dataset could support.
Columns
- Year (date)
- Country Name (string)
- Country Code (string)
- Value (integer)