Maybe-Heisenberg-07/POLARIS-Gov

POLARIS-Gov: Political Orientation Longitudinal Analysis Resource for International States

Dataset Statistics

• —Countries Tracked: 171 Sovereign Nations (160 in effective modeling universe)
• —Observations: 3,373 country-year rows
• —Temporal Coverage: 2000–2019
• —Treatment Variable: ideology_score (Unified ideological score of governing coalitions)
• —Outcome Variables: GDP growth, GDP per capita growth
• —Control Variables: 18 World Bank Economic Indicators
• —Panel Type: Unbalanced Longitudinal Panel
• —Missing Ideology Rows: 47 early-panel observations explicitly un-imputed for analytical integrity.

Quick Start

1. Using Hugging Face Datasets

from datasets import load_dataset
dataset = load_dataset("Maybe-Heisenberg-07/POLARIS-Gov")
print(dataset)

2. Using Pandas

import pandas as pd
df = pd.read_csv("hf://datasets/Maybe-Heisenberg-07/POLARIS-Gov/data/POLARIS-Gov.csv")

3. Using PyArrow

import pyarrow.csv as pacsv
table = pacsv.read_csv("hf://datasets/Maybe-Heisenberg-07/POLARIS-Gov/data/POLARIS-Gov.csv")

Intended Uses

The POLARIS-Gov dataset is explicitly designed for:

• —Causal Inference: e.g., Double Machine Learning (DML) or Difference-in-Differences to estimate the economic impact of ideological shifts.
• —Policy Analysis: Evaluating the long-term effectiveness of governing coalition orientations on macroeconomic stability.
• —Explainable AI: Serving as a tabular benchmark for SHAP/LIME interpretations on complex, real-world panel data.
• —Benchmark Datasets: Tabular regression and time-series forecasting.
• —Panel Regression & Political Economy: Cross-national empirical studies.

Dataset Provenance

graph TD
    A[Raw Political Datasets] --> B(Party Harmonization)
    B --> C(Coalition Aggregation)
    C --> D[Ideology Scoring]
    D --> E(Merge with WDI)
    E --> F{Validation}
    F --> G((Final Analytic Panel))

Dataset Motivation

Measuring the true causal impact of government ideology on national economies is historically challenging due to high dimensionality, reverse causality, and data sparsity. This dataset consolidates ideology scores (synthesizing ParlGov, Manifesto Project, V-Dem, and V-Party) and merges them cleanly with World Bank World Development Indicators (WDI) to provide a unified basis for robust cross-national analysis.

Visualizations

Temporal Coverage (2000-2019)

[image]

Ideology Score Distribution

[image]

Missing Data Percentages

[image]

Correlation Matrix (Pearson)

[image]

Known Limitations & Biases

• —The income_group variable is a fixed snapshot based on the latest available World Bank designation, not a time-varying feature.
• —Certain historical states and non-sovereign entities (e.g., Taiwan, Somaliland, Zanzibar) are excluded as they lack compatible WDI economic tracking.
• —The unbalanced nature of the panel requires careful handling of lagged variables. Researchers are advised to perform exact year-to-year matching for lagged causal inference rather than structurally shifting rows.

Ethics

This dataset utilizes public institutional data and macroeconomic records. It does not contain personally identifiable information (PII). Researchers are encouraged to apply robust causal methods when asserting links between political ideologies and economic outcomes to avoid drawing unsupported, deterministic correlations.

Recommended Citation

@misc{polaris_gov_2026,
  author       = {Heisenberg, Maybe},
  title        = {POLARIS-Gov: Political Orientation Longitudinal Analysis Resource for International States},
  year         = {2026},
  publisher    = {Hugging Face},
  howpublished = {\\url{https://huggingface.co/datasets/Maybe-Heisenberg-07/POLARIS-Gov}},
}

Further Documentation

• —Schema & Variable Overview: metadata/schema.json
• —Data Dictionary: DATA_DICTIONARY.md
• —Validation Report: VALIDATION_REPORT.md