{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Working with Census microdata\n",
    "\n",
    "[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/mmann1123/pytidycensus/blob/main/examples/05_pums_data.ipynb)\n",
    "\n",
    "Introduction to Public Use Microdata Sample (PUMS) analysis.\n",
    "\n",
    "### Census API Key\n",
    "\n",
    "To use pytidycensus, you need a free API key from the US Census Bureau. Get one at: https://api.census.gov/data/key_signup.html\n",
    "\n",
    "Set your API key:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pytidycensus as tc\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "tc.set_census_api_key(\"YOUR API KEY GOES HERE\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Understanding PUMS Data\n",
    "\n",
    "PUMS provides individual-level responses that can be used for custom analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Note: PUMS functionality would be implemented in future versions\n",
    "# This is a conceptual example of what PUMS analysis might look like\n",
    "\n",
    "print(\"PUMS data analysis capabilities:\")\n",
    "print(\"- Individual-level demographic data\")\n",
    "print(\"- Custom crosstabulations\")\n",
    "print(\"- Statistical modeling on microdata\")\n",
    "print(\"- Geographic resolution: PUMAs (100K+ population)\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## PUMS Variable Dictionary\n",
    "\n",
    "PUMS has hundreds of variables describing individuals and households:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Example of what PUMS variable exploration would look like\n",
    "# pums_vars = tc.load_pums_variables(2022, survey=\"acs1\")\n",
    "# person_vars = pums_vars[pums_vars['level'] == 'person']\n",
    "# housing_vars = pums_vars[pums_vars['level'] == 'housing']\n",
    "\n",
    "print(\"Example PUMS variables:\")\n",
    "print(\"Person-level: Age, Race, Education, Income, Employment\")\n",
    "print(\"Housing-level: Bedrooms, Value, Rent, Utilities\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Survey Weights and Statistical Analysis\n",
    "\n",
    "PUMS data requires proper handling of survey weights for accurate estimates:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Conceptual example of weighted analysis\n",
    "print(\"Survey weight considerations:\")\n",
    "print(\"- Person weights (PWGTP) for person-level analysis\")\n",
    "print(\"- Housing weights (WGTP) for housing-level analysis\")\n",
    "print(\"- Replicate weights for variance estimation\")\n",
    "print(\"- Use survey packages for proper statistical inference\")"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "name": "python",
   "version": "3.9.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}