Upload folder using huggingface_hub

LICENSE.md

@@ -0,0 +1,46 @@
+# License Information
+
+## Source Data (All Public Domain or Public Access)
+
+### U.S. Census Bureau
+- American Community Survey (ACS) 2022
+- Veteran population estimates
+- Public domain (U.S. Government work)
+
+### RAND Corporation
+- State-level firearm ownership estimates (2016)
+- Published research, publicly available
+- URL: rand.org/research/gun-policy
+
+### Bureau of Alcohol, Tobacco, Firearms and Explosives (ATF)
+- Federal Firearms Licensee (FFL) counts (2023)
+- Public domain (U.S. Government work)
+
+### Centers for Disease Control and Prevention (CDC)
+- WONDER database: Underlying Cause of Death
+- Veteran and civilian suicide rates (2021)
+- Public domain (U.S. Government work)
+
+### U.S. Department of Veterans Affairs (VA)
+- PTSD prevalence and VA healthcare utilization (2022)
+- Public domain (U.S. Government work)
+
+### U.S. Department of Defense (DoD)
+- Defense Manpower Data Center (DMDC)
+- Active duty personnel and economic impact (2023)
+- Public domain (U.S. Government work)
+
+## Combined Dataset
+
+This compiled dataset is released under **CC-BY-4.0**.
+
+When using this dataset, please cite:
+
+```
+Steuber, L. (2026). US Military & Veteran Analysis by State [Dataset].
+Kaggle. https://www.kaggle.com/datasets/lucassteuber/us-military-veteran-analysis
+```
+
+## Ethical Considerations
+
+This dataset includes sensitive metrics (suicide rates, mental health indicators). Use responsibly for research and policy analysis purposes.

README.md

@@ -6,85 +6,119 @@ task_categories:
 language:
   - en
 tags:
-  - housing
-  - affordability
-  - census
+  - veterans
+  - military
+  - firearms
+  - suicide
+  - ptsd
+  - mental-health
+  - va-healthcare
+  - us-states
   - demographics
-  - rent-burden
-  - economics
-  - us-counties
-  - geospatial
+  - public-health
+pretty_name: US Military & Veteran Analysis by State
 size_categories:
-  - 1K<n<10K
-pretty_name: US Housing Affordability Crisis by County
+  - n<1K
 dataset_info:
   features:
-    - name: fips
-      dtype: string
-    - name: county_name
+    - name: NAME
       dtype: string
     - name: state
       dtype: string
-    - name: median_gross_rent
+    - name: veteran_population
+      dtype: int64
+    - name: veteran_percentage
+      dtype: float64
+    - name: active_duty_personnel
       dtype: int64
-    - name: median_household_income
+    - name: ownership_percentage
+      dtype: float64
+    - name: ffl_count
       dtype: int64
-    - name: pct_rent_burdened_30plus
+    - name: veteran_suicide_rate
       dtype: float64
-    - name: pct_rent_burdened_50plus
+    - name: civilian_suicide_rate
       dtype: float64
-    - name: rent_to_income_ratio
+    - name: veteran_risk_ratio
       dtype: float64
-    - name: affordability_category
-      dtype: string
-    - name: hours_at_min_wage_for_rent
+    - name: ptsd_prevalence_pct
+      dtype: float64
+    - name: va_utilization_pct
       dtype: float64
   splits:
     - name: train
-      num_examples: 3222
+      num_examples: 51
 ---
 
-# US Housing Affordability Crisis by County
+# US Military & Veteran Analysis by State
 
-3,222 US counties with rent burden analysis, affordability metrics, and minimum wage comparisons.
+State-level integration of veteran demographics, firearm ownership, mental health indicators, and VA healthcare utilization from 6 authoritative sources.
 
-## Dataset Description
+## What's Inside
 
-State-level integration of housing affordability metrics from Census ACS 2022, computing rent burden percentages, income ratios, and hours at minimum wage needed for rent.
+- 51 records (50 states + DC)
+- Veteran population and percentage
+- Firearm ownership rates (RAND 2016)
+- Suicide rates: veteran vs civilian
+- PTSD prevalence estimates
+- VA healthcare utilization
 
-### Data Sources
+## Why This Dataset Exists
 
-| Metric | Source | Table |
-|--------|--------|-------|
-| Rent burden | Census ACS 2022 | B25070 |
-| Median rent | Census ACS 2022 | B25064 |
-| Median income | Census ACS 2022 | B19013 |
-| Housing tenure | Census ACS 2022 | B25003 |
+No single source provides this combined view. This integration allows correlation analysis across:
 
-### Computed Fields
+- Veteran demographics and military presence
+- Mental health indicators (suicide, PTSD)
+- Firearm access and ownership
+- Healthcare system utilization
 
-- `rent_to_income_ratio`: Monthly rent / monthly income
-- `affordability_category`: Affordable (<30%), Burdened (30-50%), Severely Burdened (>50%)
-- `hours_at_min_wage_for_rent`: Hours at $7.25/hr needed for monthly rent
+## Data Sources
 
-## Dataset Structure
+| Data | Source | Year |
+|------|--------|------|
+| Veteran population | Census ACS | 2022 |
+| Firearm ownership | RAND Corporation | 2016 |
+| FFL counts | ATF | 2023 |
+| Suicide rates | CDC WONDER | 2021 |
+| PTSD, VA utilization | VA | 2022 |
+| Active duty, economic impact | DoD DMDC | 2023 |
 
-- **Records**: 3,222 counties
-- **Format**: CSV
-- **Key**: 5-digit FIPS county code
+## Key Metrics
+
+| Field | Description |
+|-------|-------------|
+| veteran_risk_ratio | Veteran suicide rate / civilian rate |
+| ptsd_prevalence_pct | PTSD prevalence among veterans |
+| va_utilization_pct | % of veterans using VA healthcare |
+| ffl_per_100k | Federal Firearms Licensees per 100K population |
 
 ## Usage
 
 ```python
-from datasets import load_dataset
-dataset = load_dataset("lukeslp/us-housing-affordability-crisis")
-df = dataset['train'].to_pandas()
+import pandas as pd
+
+df = pd.read_csv('military_firearm_merged_analysis.csv')
+print(f"States: {len(df)}")
+
+# States with highest veteran risk ratio
+high_risk = df.nlargest(10, 'veteran_risk_ratio')[['NAME', 'veteran_risk_ratio']]
+print(high_risk)
+
+# Correlation between firearm ownership and veteran suicide
+correlation = df['ownership_percentage'].corr(df['veteran_suicide_rate'])
+print(f"Firearm-suicide correlation: {correlation:.3f}")
 ```
 
+## Limitations
+
+- RAND firearm data is from 2016 (most recent publicly available)
+- CDC suppresses data for states with <10 deaths
+- VA data covers enrolled veterans only
+
 ## License
 
-CC-BY-4.0
+CC-BY-4.0 (All source data is public domain; integration by Luke Steuber)
 
 ## Author
 
-Luke Steuber | luke@lukesteuber.com
+Luke Steuber | luke@lukesteuber.com | @lukesteuber.com (Bluesky)

dataset-metadata.json

@@ -1,13 +1,10 @@
 {
   "title": "US Military & Veteran Analysis by State",
   "id": "lucassteuber/us-military-veteran-analysis",
-  "licenses": [
-    {
-      "name": "CC-BY-4.0"
-    }
-  ],
-  "subtitle": "50 states: veterans, firearms, PTSD, suicide, VA healthcare integrated",
-  "description": "# US Military & Veteran Analysis by State\n\nState-level integration of veteran demographics, firearm ownership, mental health indicators, and VA healthcare utilization from **6 independent government and research sources**.\n\n## What Makes This Dataset Unique\n\nNo single source provides this combined view. This dataset integrates:\n\n1. **Census ACS** - Veteran population, total population\n2. **RAND Corporation** - State firearm ownership rates (2016)\n3. **ATF** - Federal Firearms Licensee (FFL) counts\n4. **CDC** - Veteran and civilian suicide rates, risk ratios\n5. **VA** - PTSD prevalence, healthcare utilization\n6. **DoD** - Active duty personnel, military installation economic impact\n\n## Data Sources\n\n| Source | Organization | Access |\n|--------|--------------|--------|\n| American Community Survey | U.S. Census Bureau | api.census.gov |\n| Firearm Ownership Estimates | RAND Corporation | rand.org |\n| Federal Firearms Licensees | ATF | atf.gov |\n| Underlying Cause of Death | CDC WONDER | wonder.cdc.gov |\n| Veterans Health Administration | U.S. Dept of Veterans Affairs | va.gov |\n| Defense Manpower Data | DoD DMDC | dmdc.osd.mil |\n\n## Fields\n\n- `NAME`: State name\n- `state`: State FIPS code\n- `veteran_population`: Total veteran population\n- `veteran_percentage`: Veterans as % of total population\n- `active_duty_personnel`: Active duty military personnel\n- `ownership_percentage`: Household firearm ownership rate (RAND 2016)\n- `ffl_count`: Federal Firearms Licensees in state\n- `ffl_per_100k`: FFLs per 100,000 population\n- `veteran_suicide_rate`: Veteran suicide rate per 100,000\n- `civilian_suicide_rate`: Civilian suicide rate per 100,000\n- `veteran_risk_ratio`: Veteran/civilian suicide risk ratio\n- `ptsd_prevalence_pct`: PTSD prevalence among veterans\n- `va_utilization_pct`: Percent of veterans using VA healthcare\n- `annual_economic_impact_millions`: Military installation economic impact\n\n## Records\n\n54 (50 states + DC + territories)\n\n## Limitations\n\n- RAND firearm data is from 2016 (most recent available)\n- CDC suppresses counties with <10 deaths\n- VA data covers enrolled veterans only\n\n## License\n\nCC-BY-4.0 (All source data is public domain; integration by Luke Steuber)",
+  "licenses": [{"name": "CC-BY-4.0"}],
+  "subtitle": "50 states: veterans, firearms, PTSD, suicide rates, VA healthcare",
+  "description": "State-level integration of veteran demographics, firearm ownership, mental health indicators, and VA healthcare utilization from Census ACS, RAND, ATF, CDC, VA, and DoD.",
+  "isPrivate": false,
   "keywords": [
     "veterans",
     "military",
@@ -19,6 +16,5 @@
     "us-states",
     "demographics",
     "public-health"
-  ],
-  "isPrivate": true
+  ]
 }

demo_notebook.ipynb

@@ -0,0 +1,263 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "# US Military & Veteran Analysis - Interactive Demo\n",
+        "\n",
+        "Explore **state-level data** on veteran demographics, firearm ownership, mental health indicators, and VA healthcare.\n",
+        "\n",
+        "**Dataset Highlights:**\n",
+        "- 6 authoritative sources integrated\n",
+        "- Veteran vs civilian suicide rates\n",
+        "- PTSD prevalence estimates\n",
+        "- VA healthcare utilization"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "import pandas as pd\n",
+        "import numpy as np\n",
+        "import matplotlib.pyplot as plt\n",
+        "import seaborn as sns\n",
+        "import warnings\n",
+        "warnings.filterwarnings('ignore')\n",
+        "\n",
+        "plt.style.use('seaborn-v0_8-darkgrid')\n",
+        "sns.set_palette('husl')\n",
+        "\n",
+        "print('Libraries loaded')"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 1. Load Dataset"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "df = pd.read_csv('military_firearm_merged_analysis.csv')\n",
+        "print(f'Total states: {len(df)}')\n",
+        "df.head()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 2. Dataset Overview"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "print('Dataset Statistics:')\n",
+        "print('=' * 60)\n",
+        "print(f\"States: {len(df)}\")\n",
+        "print(f\"\\nVeteran Statistics:\")\n",
+        "print(f\"  Total veteran population: {df['veteran_population'].sum():,.0f}\")\n",
+        "print(f\"  Avg veteran %: {df['veteran_percentage'].mean():.1f}%\")\n",
+        "print(f\"\\nMental Health:\")\n",
+        "print(f\"  Avg veteran suicide rate: {df['veteran_suicide_rate'].mean():.1f} per 100K\")\n",
+        "print(f\"  Avg civilian suicide rate: {df['civilian_suicide_rate'].mean():.1f} per 100K\")\n",
+        "print(f\"  Avg veteran risk ratio: {df['veteran_risk_ratio'].mean():.2f}x\")"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 3. Veteran Risk Analysis"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "# Top 10 states by veteran risk ratio\n",
+        "high_risk = df.nlargest(10, 'veteran_risk_ratio')[['NAME', 'veteran_risk_ratio', 'veteran_suicide_rate', 'civilian_suicide_rate']]\n",
+        "print('States with Highest Veteran Risk Ratio:')\n",
+        "print(high_risk.to_string(index=False))"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "# Compare veteran vs civilian suicide rates\n",
+        "fig, ax = plt.subplots(figsize=(12, 8))\n",
+        "\n",
+        "sorted_df = df.sort_values('veteran_suicide_rate', ascending=True).tail(20)\n",
+        "y_pos = range(len(sorted_df))\n",
+        "\n",
+        "ax.barh(y_pos, sorted_df['veteran_suicide_rate'], height=0.4, label='Veteran', color='steelblue', alpha=0.8)\n",
+        "ax.barh([y + 0.4 for y in y_pos], sorted_df['civilian_suicide_rate'], height=0.4, label='Civilian', color='coral', alpha=0.8)\n",
+        "\n",
+        "ax.set_yticks([y + 0.2 for y in y_pos])\n",
+        "ax.set_yticklabels(sorted_df['NAME'])\n",
+        "ax.set_xlabel('Suicide Rate per 100,000', fontsize=12)\n",
+        "ax.set_title('Veteran vs Civilian Suicide Rates by State', fontsize=14, fontweight='bold')\n",
+        "ax.legend()\n",
+        "\n",
+        "plt.tight_layout()\n",
+        "plt.show()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 4. Firearm Ownership Analysis"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "print('Firearm Ownership Statistics:')\n",
+        "print('=' * 50)\n",
+        "print(f\"Mean ownership rate: {df['ownership_percentage'].mean():.1f}%\")\n",
+        "print(f\"Highest: {df.loc[df['ownership_percentage'].idxmax(), 'NAME']} ({df['ownership_percentage'].max():.1f}%)\")\n",
+        "print(f\"Lowest: {df.loc[df['ownership_percentage'].idxmin(), 'NAME']} ({df['ownership_percentage'].min():.1f}%)\")"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "# Correlation between firearm ownership and veteran suicide\n",
+        "fig, ax = plt.subplots(figsize=(10, 8))\n",
+        "\n",
+        "ax.scatter(df['ownership_percentage'], df['veteran_suicide_rate'], alpha=0.7, s=100)\n",
+        "\n",
+        "# Add correlation line\n",
+        "z = np.polyfit(df['ownership_percentage'], df['veteran_suicide_rate'], 1)\n",
+        "p = np.poly1d(z)\n",
+        "ax.plot(df['ownership_percentage'].sort_values(), p(df['ownership_percentage'].sort_values()), \n",
+        "        'r--', alpha=0.8, label=f'r = {df[\"ownership_percentage\"].corr(df[\"veteran_suicide_rate\"]):.3f}')\n",
+        "\n",
+        "ax.set_xlabel('Firearm Ownership Rate (%)', fontsize=12)\n",
+        "ax.set_ylabel('Veteran Suicide Rate (per 100K)', fontsize=12)\n",
+        "ax.set_title('Firearm Ownership vs Veteran Suicide Rate', fontsize=14, fontweight='bold')\n",
+        "ax.legend()\n",
+        "\n",
+        "plt.tight_layout()\n",
+        "plt.show()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 5. VA Healthcare Utilization"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "print('VA Healthcare Utilization:')\n",
+        "print('=' * 50)\n",
+        "print(f\"Mean utilization: {df['va_utilization_pct'].mean():.1f}%\")\n",
+        "print(f\"\\nTop 5 States (highest VA use):\")\n",
+        "top_va = df.nlargest(5, 'va_utilization_pct')[['NAME', 'va_utilization_pct', 'veteran_population']]\n",
+        "print(top_va.to_string(index=False))"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "# PTSD prevalence by state\n",
+        "if 'ptsd_prevalence_pct' in df.columns:\n",
+        "    fig, ax = plt.subplots(figsize=(14, 6))\n",
+        "    sorted_ptsd = df.sort_values('ptsd_prevalence_pct')\n",
+        "    ax.barh(sorted_ptsd['NAME'], sorted_ptsd['ptsd_prevalence_pct'], color='teal')\n",
+        "    ax.set_xlabel('PTSD Prevalence (%)', fontsize=12)\n",
+        "    ax.set_title('PTSD Prevalence Among Veterans by State', fontsize=14, fontweight='bold')\n",
+        "    plt.tight_layout()\n",
+        "    plt.show()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## 6. Correlation Analysis"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "# Key correlations\n",
+        "metrics = ['veteran_suicide_rate', 'ownership_percentage', 'ptsd_prevalence_pct', 'va_utilization_pct', 'veteran_percentage']\n",
+        "corr_matrix = df[metrics].corr()\n",
+        "\n",
+        "fig, ax = plt.subplots(figsize=(10, 8))\n",
+        "sns.heatmap(corr_matrix, annot=True, cmap='RdBu_r', center=0, ax=ax, fmt='.2f')\n",
+        "ax.set_title('Correlation Matrix', fontsize=14, fontweight='bold')\n",
+        "plt.tight_layout()\n",
+        "plt.show()"
+      ]
+    },
+    {
+      "cell_type": "markdown",
+      "metadata": {},
+      "source": [
+        "## Conclusion\n",
+        "\n",
+        "This notebook demonstrated:\n",
+        "- Loading and exploring veteran analysis data\n",
+        "- Comparing veteran vs civilian suicide rates\n",
+        "- Analyzing firearm ownership correlations\n",
+        "- VA healthcare utilization patterns\n",
+        "\n",
+        "**Important Note**: RAND firearm data is from 2016 (most recent publicly available). Use findings as exploratory, not causal.\n",
+        "\n",
+        "**Author**: Luke Steuber | luke@lukesteuber.com | @lukesteuber.com (Bluesky)"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "name": "python",
+      "version": "3.10.0"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 4
+}