{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Imprisonment by Race\n",
    "\n",
    "This notebook explores the racial breakdown of the 3 largest racial groups in US prisons (from 2006 to 2016): non-Hispanic white people, non-Hispanic black people, and Hispanic people. \n",
    "\n",
    "In this notebook, I show that, over the entire span of data (2006-2016), black people are the largest racial prison population both in raw numbers and as a proportion of the total same-race US population. \n",
    "\n",
    "**Average Number of Prisoners (from 2006-2016)**\n",
    "\n",
    "| Race | Average Number in Prison | \n",
    "| ------------- |:-------------:| \n",
    "| Black | 551209 | \n",
    "| White | 476136 | \n",
    "| Hispanic | 336500 | \n",
    "\n",
    "**Average Percent US Racial Population in Prison**\n",
    "\n",
    "| Race | % of US Racial Pop. in Prison | \n",
    "| ------------- | :------------- | \n",
    "| Black | 1.440 % |\n",
    "| White | 0.241 % |\n",
    "| Hispanic | 0.656 % |\n",
    "\n",
    "**Change in Average Percent US Racial Population in Prison**\n",
    "\n",
    "| Race | Change in % of US Racial Pop. in Prison | \n",
    "| ------------- | :------------- | \n",
    "| Black | -0.406 % |\n",
    "| White | -0.035 % |\n",
    "| Hispanic | -0.113 % |\n",
    "\n",
    "That is a fairly astonishing. A randomly selected black person is (on average) nearly 6 times as likely to be in prison as a randomly selected white person. There is clearly a statistically significant difference between these rates, so there are clearly more questions to answer. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Source\n",
    "\n",
    "This data is from file p16t03.csv of the Bureau of Justice Statistics [Prisoner Series](https://www.bjs.gov/index.cfm?ty=pbse&sid=40) data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as ticker\n",
    "import os\n",
    "from IPython.core.display import display, HTML\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Importing modules from a visualization package.\n",
    "# from bokeh.sampledata.us_states import data as states|\n",
    "from bokeh.plotting import figure, show, output_notebook\n",
    "from bokeh.models import HoverTool, ColumnDataSource\n",
    "from bokeh.models import LinearColorMapper, ColorBar, BasicTicker"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<style>.container { width:100% !important; }</style>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# styling\n",
    "pd.options.display.max_columns = None\n",
    "display(HTML(\"<style>.container { width:100% !important; }</style>\"))\n",
    "pd.set_option('display.float_format',lambda x: '%.3f' % x)\n",
    "plt.rcParams['figure.figsize'] = 10,10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "CSV_PATH = os.path.join('data', 'prison', 'p16t03.csv')\n",
    "race_sex_raw = pd.read_csv(CSV_PATH, \n",
    "                           encoding='latin1',\n",
    "                           header=11, \n",
    "                           na_values=':',\n",
    "                           thousands=r',')\n",
    "race_sex_raw.dropna(axis=0, thresh=3, inplace=True)\n",
    "race_sex_raw.dropna(axis=1, thresh=3, inplace=True)\n",
    "race_sex_raw.dropna(axis=0, inplace=True)\n",
    "fix = lambda x: x.split('/')[0]\n",
    "race_sex_raw['Year'] = race_sex_raw['Year'].apply(fix)\n",
    "race_sex_raw.columns = [x.split('/')[0] for x in race_sex_raw.columns]\n",
    "race_sex_raw.set_index('Year', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Total</th>\n",
       "      <th>Federal</th>\n",
       "      <th>State</th>\n",
       "      <th>Male</th>\n",
       "      <th>Female</th>\n",
       "      <th>White</th>\n",
       "      <th>Black</th>\n",
       "      <th>Hispanic</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Year</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2006</th>\n",
       "      <td>1504598.0</td>\n",
       "      <td>173533.0</td>\n",
       "      <td>1331065.0</td>\n",
       "      <td>1401261.0</td>\n",
       "      <td>103337.0</td>\n",
       "      <td>507100.0</td>\n",
       "      <td>590300.0</td>\n",
       "      <td>313600.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>1532851.0</td>\n",
       "      <td>179204.0</td>\n",
       "      <td>1353647.0</td>\n",
       "      <td>1427088.0</td>\n",
       "      <td>105763.0</td>\n",
       "      <td>499800.0</td>\n",
       "      <td>592900.0</td>\n",
       "      <td>330400.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008</th>\n",
       "      <td>1547742.0</td>\n",
       "      <td>182333.0</td>\n",
       "      <td>1365409.0</td>\n",
       "      <td>1441384.0</td>\n",
       "      <td>106358.0</td>\n",
       "      <td>499900.0</td>\n",
       "      <td>592800.0</td>\n",
       "      <td>329800.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009</th>\n",
       "      <td>1553574.0</td>\n",
       "      <td>187886.0</td>\n",
       "      <td>1365688.0</td>\n",
       "      <td>1448239.0</td>\n",
       "      <td>105335.0</td>\n",
       "      <td>490000.0</td>\n",
       "      <td>584800.0</td>\n",
       "      <td>341200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010</th>\n",
       "      <td>1552669.0</td>\n",
       "      <td>190641.0</td>\n",
       "      <td>1362028.0</td>\n",
       "      <td>1447766.0</td>\n",
       "      <td>104903.0</td>\n",
       "      <td>484400.0</td>\n",
       "      <td>572700.0</td>\n",
       "      <td>345800.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011</th>\n",
       "      <td>1538847.0</td>\n",
       "      <td>197050.0</td>\n",
       "      <td>1341797.0</td>\n",
       "      <td>1435141.0</td>\n",
       "      <td>103706.0</td>\n",
       "      <td>474300.0</td>\n",
       "      <td>557100.0</td>\n",
       "      <td>347800.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012</th>\n",
       "      <td>1512430.0</td>\n",
       "      <td>196574.0</td>\n",
       "      <td>1315856.0</td>\n",
       "      <td>1411076.0</td>\n",
       "      <td>101354.0</td>\n",
       "      <td>466600.0</td>\n",
       "      <td>537800.0</td>\n",
       "      <td>340300.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013</th>\n",
       "      <td>1520403.0</td>\n",
       "      <td>195098.0</td>\n",
       "      <td>1325305.0</td>\n",
       "      <td>1416102.0</td>\n",
       "      <td>104301.0</td>\n",
       "      <td>463900.0</td>\n",
       "      <td>529900.0</td>\n",
       "      <td>341200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014</th>\n",
       "      <td>1507781.0</td>\n",
       "      <td>191374.0</td>\n",
       "      <td>1316407.0</td>\n",
       "      <td>1401685.0</td>\n",
       "      <td>106096.0</td>\n",
       "      <td>461500.0</td>\n",
       "      <td>518700.0</td>\n",
       "      <td>338900.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015</th>\n",
       "      <td>1476847.0</td>\n",
       "      <td>178688.0</td>\n",
       "      <td>1298159.0</td>\n",
       "      <td>1371879.0</td>\n",
       "      <td>104968.0</td>\n",
       "      <td>450200.0</td>\n",
       "      <td>499400.0</td>\n",
       "      <td>333200.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016</th>\n",
       "      <td>1458173.0</td>\n",
       "      <td>171482.0</td>\n",
       "      <td>1286691.0</td>\n",
       "      <td>1352684.0</td>\n",
       "      <td>105489.0</td>\n",
       "      <td>439800.0</td>\n",
       "      <td>486900.0</td>\n",
       "      <td>339300.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Total   Federal      State       Male    Female     White     Black  \\\n",
       "Year                                                                            \n",
       "2006  1504598.0  173533.0  1331065.0  1401261.0  103337.0  507100.0  590300.0   \n",
       "2007  1532851.0  179204.0  1353647.0  1427088.0  105763.0  499800.0  592900.0   \n",
       "2008  1547742.0  182333.0  1365409.0  1441384.0  106358.0  499900.0  592800.0   \n",
       "2009  1553574.0  187886.0  1365688.0  1448239.0  105335.0  490000.0  584800.0   \n",
       "2010  1552669.0  190641.0  1362028.0  1447766.0  104903.0  484400.0  572700.0   \n",
       "2011  1538847.0  197050.0  1341797.0  1435141.0  103706.0  474300.0  557100.0   \n",
       "2012  1512430.0  196574.0  1315856.0  1411076.0  101354.0  466600.0  537800.0   \n",
       "2013  1520403.0  195098.0  1325305.0  1416102.0  104301.0  463900.0  529900.0   \n",
       "2014  1507781.0  191374.0  1316407.0  1401685.0  106096.0  461500.0  518700.0   \n",
       "2015  1476847.0  178688.0  1298159.0  1371879.0  104968.0  450200.0  499400.0   \n",
       "2016  1458173.0  171482.0  1286691.0  1352684.0  105489.0  439800.0  486900.0   \n",
       "\n",
       "      Hispanic  \n",
       "Year            \n",
       "2006  313600.0  \n",
       "2007  330400.0  \n",
       "2008  329800.0  \n",
       "2009  341200.0  \n",
       "2010  345800.0  \n",
       "2011  347800.0  \n",
       "2012  340300.0  \n",
       "2013  341200.0  \n",
       "2014  338900.0  \n",
       "2015  333200.0  \n",
       "2016  339300.0  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "race_sex_raw"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average number of    black people in prison from 2006 to 2016: 551209\n",
      "Average number of    White people in prison from 2006 to 2016: 476136\n",
      "Average number of Hispanic people in prison from 2006 to 2016: 336500\n"
     ]
    }
   ],
   "source": [
    "print('Average number of {:>8s} people in prison from 2006 to 2016: {:6.0f}'\n",
    "      .format('black', race_sex_raw['Black'].mean()))\n",
    "print('Average number of {:>8s} people in prison from 2006 to 2016: {:6.0f}'\n",
    "      .format('White', race_sex_raw['White'].mean()))\n",
    "print('Average number of {:>8s} people in prison from 2006 to 2016: {:6.0f}'\n",
    "      .format('Hispanic', race_sex_raw['Hispanic'].mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x244fb7b2f28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.axes_style(\"whitegrid\"):\n",
    "    fig, ax = plt.subplots(figsize=(10,7))\n",
    "    ax.plot(race_sex_raw['White'])\n",
    "    ax.plot(race_sex_raw['Black'])\n",
    "    ax.plot(race_sex_raw['Hispanic'])\n",
    "    ax.set_title('Total Imprisonment Counts by Race (table: p16t03)',fontsize=14)\n",
    "    ax.set_xlabel('Year', fontsize=14)\n",
    "    ax.set_ylabel('Number of People imprisoned', fontsize=14)\n",
    "    ax.legend(fontsize=14)\n",
    "    ax.set_ylim([0, 1.1*max([race_sex_raw['White'].max(), \n",
    "                             race_sex_raw['Black'].max(),\n",
    "                             race_sex_raw['Hispanic'].max()])])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looking at this plot of total imprisonment counts, we see that:\n",
    "* At any given time, there are more non-Hispanic black people in prison than any other race.\n",
    "* At any given time, there are more non-Hispanic white people than Hispanic people in prison.\n",
    "* The numbers of imprisoned black people and white people decreased steadily from 2008 to 2016, while the number of imprisoned Hispanic people was been fairly flat over that time.\n",
    "\n",
    "These numbers are raw counts, so they don't account for the fact that the fact that these races make up different proportions of the entire US population. This motivates questions like:\n",
    "\n",
    "* What percent of the population of each race is in prison?\n",
    "\n",
    "To answer that, I'll have to find population data broken down by race.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Census Data\n",
    "\n",
    "The Constitution requires that a full, national census is performed every 10 years that reaches every resident on American soil. This data is used to determine how much federal money is allocated to each district for things like schools, roadways, police, etc. and it determines how many congressional representatives will be allocated to each state. With the help of smaller surveys, the US Census bureau makes estimates of populations for the years between censuses.\n",
    "\n",
    "https://factfinder.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Initial Data Format:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Id</th>\n",
       "      <th>Year</th>\n",
       "      <th>Id.1</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Id.2</th>\n",
       "      <th>Hispanic Origin</th>\n",
       "      <th>Id.3</th>\n",
       "      <th>Id2</th>\n",
       "      <th>Geography</th>\n",
       "      <th>Total</th>\n",
       "      <th>Race Alone - White</th>\n",
       "      <th>Race Alone - Black or African American</th>\n",
       "      <th>Race Alone - American Indian and Alaska Native</th>\n",
       "      <th>Race Alone - Asian</th>\n",
       "      <th>Race Alone - Native Hawaiian and Other Pacific Islander</th>\n",
       "      <th>Two or More Races</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>cen42010</td>\n",
       "      <td>April 1, 2010 Census</td>\n",
       "      <td>female</td>\n",
       "      <td>Female</td>\n",
       "      <td>hisp</td>\n",
       "      <td>Hispanic</td>\n",
       "      <td>0100000US</td>\n",
       "      <td>NaN</td>\n",
       "      <td>United States</td>\n",
       "      <td>24858794</td>\n",
       "      <td>21936806</td>\n",
       "      <td>1191984</td>\n",
       "      <td>702309</td>\n",
       "      <td>249346</td>\n",
       "      <td>85203</td>\n",
       "      <td>693146</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>cen42010</td>\n",
       "      <td>April 1, 2010 Census</td>\n",
       "      <td>female</td>\n",
       "      <td>Female</td>\n",
       "      <td>nhisp</td>\n",
       "      <td>Not Hispanic</td>\n",
       "      <td>0100000US</td>\n",
       "      <td>NaN</td>\n",
       "      <td>United States</td>\n",
       "      <td>132105418</td>\n",
       "      <td>100301335</td>\n",
       "      <td>19853611</td>\n",
       "      <td>1147502</td>\n",
       "      <td>7691693</td>\n",
       "      <td>246518</td>\n",
       "      <td>2864759</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>cen42010</td>\n",
       "      <td>April 1, 2010 Census</td>\n",
       "      <td>female</td>\n",
       "      <td>Female</td>\n",
       "      <td>tothisp</td>\n",
       "      <td>Total</td>\n",
       "      <td>0100000US</td>\n",
       "      <td>NaN</td>\n",
       "      <td>United States</td>\n",
       "      <td>156964212</td>\n",
       "      <td>122238141</td>\n",
       "      <td>21045595</td>\n",
       "      <td>1849811</td>\n",
       "      <td>7941039</td>\n",
       "      <td>331721</td>\n",
       "      <td>3557905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>cen42010</td>\n",
       "      <td>April 1, 2010 Census</td>\n",
       "      <td>male</td>\n",
       "      <td>Male</td>\n",
       "      <td>hisp</td>\n",
       "      <td>Hispanic</td>\n",
       "      <td>0100000US</td>\n",
       "      <td>NaN</td>\n",
       "      <td>United States</td>\n",
       "      <td>25618800</td>\n",
       "      <td>22681299</td>\n",
       "      <td>1136129</td>\n",
       "      <td>773939</td>\n",
       "      <td>248654</td>\n",
       "      <td>92206</td>\n",
       "      <td>686573</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>cen42010</td>\n",
       "      <td>April 1, 2010 Census</td>\n",
       "      <td>male</td>\n",
       "      <td>Male</td>\n",
       "      <td>nhisp</td>\n",
       "      <td>Not Hispanic</td>\n",
       "      <td>0100000US</td>\n",
       "      <td>NaN</td>\n",
       "      <td>United States</td>\n",
       "      <td>126162526</td>\n",
       "      <td>97017621</td>\n",
       "      <td>18068911</td>\n",
       "      <td>1115756</td>\n",
       "      <td>6969823</td>\n",
       "      <td>250698</td>\n",
       "      <td>2739717</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Id                  Year    Id.1     Sex     Id.2 Hispanic Origin  \\\n",
       "0  cen42010  April 1, 2010 Census  female  Female     hisp        Hispanic   \n",
       "1  cen42010  April 1, 2010 Census  female  Female    nhisp    Not Hispanic   \n",
       "2  cen42010  April 1, 2010 Census  female  Female  tothisp           Total   \n",
       "3  cen42010  April 1, 2010 Census    male    Male     hisp        Hispanic   \n",
       "4  cen42010  April 1, 2010 Census    male    Male    nhisp    Not Hispanic   \n",
       "\n",
       "        Id.3  Id2      Geography      Total  Race Alone - White  \\\n",
       "0  0100000US  NaN  United States   24858794            21936806   \n",
       "1  0100000US  NaN  United States  132105418           100301335   \n",
       "2  0100000US  NaN  United States  156964212           122238141   \n",
       "3  0100000US  NaN  United States   25618800            22681299   \n",
       "4  0100000US  NaN  United States  126162526            97017621   \n",
       "\n",
       "   Race Alone - Black or African American  \\\n",
       "0                                 1191984   \n",
       "1                                19853611   \n",
       "2                                21045595   \n",
       "3                                 1136129   \n",
       "4                                18068911   \n",
       "\n",
       "   Race Alone - American Indian and Alaska Native  Race Alone - Asian  \\\n",
       "0                                          702309              249346   \n",
       "1                                         1147502             7691693   \n",
       "2                                         1849811             7941039   \n",
       "3                                          773939              248654   \n",
       "4                                         1115756             6969823   \n",
       "\n",
       "   Race Alone - Native Hawaiian and Other Pacific Islander  Two or More Races  \n",
       "0                                              85203                   693146  \n",
       "1                                             246518                  2864759  \n",
       "2                                             331721                  3557905  \n",
       "3                                              92206                   686573  \n",
       "4                                             250698                  2739717  "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Data Format after Processing:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Year</th>\n",
       "      <th>Hispanic Origin</th>\n",
       "      <th>Total</th>\n",
       "      <th>white_only_pop</th>\n",
       "      <th>black_only_pop</th>\n",
       "      <th>Two or More Races</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>July 1, 2010</td>\n",
       "      <td>Hispanic</td>\n",
       "      <td>50754069</td>\n",
       "      <td>44855529</td>\n",
       "      <td>2343053</td>\n",
       "      <td>1392607</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>July 1, 2010</td>\n",
       "      <td>Not Hispanic</td>\n",
       "      <td>258594124</td>\n",
       "      <td>197394319</td>\n",
       "      <td>38015461</td>\n",
       "      <td>5647760</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>July 1, 2011</td>\n",
       "      <td>Hispanic</td>\n",
       "      <td>51906353</td>\n",
       "      <td>45841771</td>\n",
       "      <td>2410490</td>\n",
       "      <td>1445820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>July 1, 2011</td>\n",
       "      <td>Not Hispanic</td>\n",
       "      <td>259757005</td>\n",
       "      <td>197519026</td>\n",
       "      <td>38393758</td>\n",
       "      <td>5826406</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>July 1, 2012</td>\n",
       "      <td>Hispanic</td>\n",
       "      <td>52993496</td>\n",
       "      <td>46759165</td>\n",
       "      <td>2480193</td>\n",
       "      <td>1498388</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            Year Hispanic Origin      Total  white_only_pop  black_only_pop  \\\n",
       "24  July 1, 2010        Hispanic   50754069        44855529         2343053   \n",
       "25  July 1, 2010    Not Hispanic  258594124       197394319        38015461   \n",
       "33  July 1, 2011        Hispanic   51906353        45841771         2410490   \n",
       "34  July 1, 2011    Not Hispanic  259757005       197519026        38393758   \n",
       "42  July 1, 2012        Hispanic   52993496        46759165         2480193   \n",
       "\n",
       "    Two or More Races  \n",
       "24            1392607  \n",
       "25            5647760  \n",
       "33            1445820  \n",
       "34            5826406  \n",
       "42            1498388  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "CSV_PATH = os.path.join('data', 'pop', 'PEP_2016_PEPSR6H_with_ann.csv')\n",
    "race_pop = pd.read_csv(CSV_PATH, header=[1], encoding='latin1')\n",
    "print('Initial Data Format:')\n",
    "display(race_pop.head())\n",
    "# Only looking for national values\n",
    "race_pop = race_pop[race_pop['Geography'] == 'United States']\n",
    "# Eliminating the actual census values\n",
    "race_pop = race_pop[~race_pop['Year'].str.contains('April')]\n",
    "# Eliminating aggregated rows\n",
    "race_pop = race_pop[race_pop['Hispanic Origin'] != 'Total']\n",
    "# race_pop = race_pop[race_pop['Sex'] != 'Both Sexes']  # for later\n",
    "race_pop = race_pop[~race_pop['Sex'].isin(['Male','Female'])]\n",
    "\n",
    "drop_cols = ['Id', 'Id.1', 'Id.2', 'Id2', 'Id.3', 'Geography', 'Sex',\n",
    "             'Race Alone - American Indian and Alaska Native', 'Race Alone - Asian',\n",
    "             'Race Alone - Native Hawaiian and Other Pacific Islander']\n",
    "race_pop.drop(drop_cols, axis=1, inplace=True)\n",
    "\n",
    "# Reducing the size of long column names\n",
    "col_map = {'Race Alone - Black or African American':'black_only_pop',\n",
    "           'Race Alone - White': 'white_only_pop'}\n",
    "race_pop.rename(col_map, axis=1, inplace=True)\n",
    "\n",
    "print('\\nData Format after Processing:')\n",
    "race_pop.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the aggregation of the data I pulled, there is a rather vague 'Two or More Races' column. The prisoner data I've been working with has not differentiated between multiracial people and single-race people, rather, the prison data only breaks races down to non-Hispanic white, non-Hispanic black, and Hispanic. It's entirely possible for someone to be non-Hispanic, black, and multiracial, which would make it much more difficult and less accurate to use this population data with the prison data. Fortunately, however, the [documentation for the prison data](https://www.bjs.gov/content/pub/pdf/p16.pd)] includes footnotes indicating the data for white and black prison populations excludes persons of two or more races, so, conveniently, I must also exclude it."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "race_pop.drop('Two or More Races', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "hisp = race_pop[race_pop['Hispanic Origin'] == 'Hispanic'].copy()\n",
    "non_hisp = race_pop[race_pop['Hispanic Origin'] != 'Hispanic'].copy()\n",
    "non_hisp.drop(['Hispanic Origin', 'Total'], axis=1, inplace=True)\n",
    "hisp.drop(['white_only_pop','black_only_pop', 'Hispanic Origin'], axis=1, inplace=True)\n",
    "hisp.rename({'Total':'Hispanic_pop'}, axis=1, inplace=True)\n",
    "us_race_pop = hisp.merge(non_hisp, on=['Year'])\n",
    "fix_yr = lambda x: x.split(' ')[-1]\n",
    "us_race_pop['Year'] = us_race_pop['Year'].apply(fix_yr)\n",
    "us_race_pop.set_index('Year', inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This provides population data from 2010 on, but not for the earlier years. To deal with earlier years, I need to handle another data set. \n",
    "\n",
    "https://www.census.gov/data/tables/time-series/demo/popest/intercensal-2000-2010-national.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "EXCEL_PATH = os.path.join('data', 'pop', 'us-est00int-02.xls')\n",
    "race_pop0010 = pd.read_excel(EXCEL_PATH, header=None)\n",
    "race_pop0010.dropna(axis=0, thresh=6, inplace=True)\n",
    "race_pop0010.drop([1],axis=1, inplace=True)\n",
    "race_pop0010 = race_pop0010.T\n",
    "race_pop0010.iloc[0,0] = 'Year'\n",
    "race_pop0010.columns = race_pop0010.loc[0]\n",
    "race_pop0010.drop(0, axis=0, inplace=True)\n",
    "race_pop0010.dropna(axis=0, inplace=True)\n",
    "race_pop0010['Year'] = race_pop0010['Year'].astype(int)\n",
    "race_pop0010['Year'] = race_pop0010['Year'].astype(str)\n",
    "race_pop0010.set_index('Year', inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Code to help drop columns I'm not interested in\n",
    "drop_cols = []\n",
    "drop_stumps = ['AIAN','Asian','NHPI', 'One Race', 'Two or']\n",
    "for col in race_pop0010.columns:\n",
    "    if any(x in col for x in drop_stumps):\n",
    "        drop_cols.append(col)\n",
    "race_pop0010.drop(drop_cols, axis=1, inplace=True)\n",
    "\n",
    "# Code to facilitate merging this DataFrame with theother Population DataFrame\n",
    "both0010 = race_pop0010.iloc[:,4:7].copy()\n",
    "name_map = {'...White'  : 'white_only_pop',\n",
    "            '...Black'  : 'black_only_pop',\n",
    "            '.HISPANIC' : 'Hispanic_pop'}\n",
    "both0010.rename(name_map, axis=1, inplace=True)\n",
    "both0010 = both0010.astype(int)\n",
    "pop_span = pd.concat([both0010, us_race_pop], join='inner')\n",
    "race_sex_pop = race_sex_raw.join(pop_span)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now that I've got a full population data set, I can normalize the prisoner data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "race_sex_pop.loc[:,'White_pct'] = race_sex_pop.loc[:,'White']\\\n",
    "                .divide(race_sex_pop.loc[:,'white_only_pop']) * 100\n",
    "race_sex_pop.loc[:,'Black_pct'] = race_sex_pop.loc[:,'Black']\\\n",
    "                .divide(race_sex_pop.loc[:,'black_only_pop']) * 100\n",
    "race_sex_pop.loc[:,'Hispanic_pct'] = race_sex_pop.loc[:,'Hispanic']\\\n",
    "                .divide(race_sex_pop.loc[:,'Hispanic_pop']) * 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Total</th>\n",
       "      <th>Federal</th>\n",
       "      <th>State</th>\n",
       "      <th>Male</th>\n",
       "      <th>Female</th>\n",
       "      <th>White</th>\n",
       "      <th>Black</th>\n",
       "      <th>Hispanic</th>\n",
       "      <th>white_only_pop</th>\n",
       "      <th>black_only_pop</th>\n",
       "      <th>Hispanic_pop</th>\n",
       "      <th>White_pct</th>\n",
       "      <th>Black_pct</th>\n",
       "      <th>Hispanic_pct</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Year</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2006</th>\n",
       "      <td>1504598.0</td>\n",
       "      <td>173533.0</td>\n",
       "      <td>1331065.0</td>\n",
       "      <td>1401261.0</td>\n",
       "      <td>103337.0</td>\n",
       "      <td>507100.0</td>\n",
       "      <td>590300.0</td>\n",
       "      <td>313600.0</td>\n",
       "      <td>196832697</td>\n",
       "      <td>36520961</td>\n",
       "      <td>44606305</td>\n",
       "      <td>0.257630</td>\n",
       "      <td>1.616332</td>\n",
       "      <td>0.703040</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>1532851.0</td>\n",
       "      <td>179204.0</td>\n",
       "      <td>1353647.0</td>\n",
       "      <td>1427088.0</td>\n",
       "      <td>105763.0</td>\n",
       "      <td>499800.0</td>\n",
       "      <td>592900.0</td>\n",
       "      <td>330400.0</td>\n",
       "      <td>197011394</td>\n",
       "      <td>36905758</td>\n",
       "      <td>46196853</td>\n",
       "      <td>0.253691</td>\n",
       "      <td>1.606524</td>\n",
       "      <td>0.715200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008</th>\n",
       "      <td>1547742.0</td>\n",
       "      <td>182333.0</td>\n",
       "      <td>1365409.0</td>\n",
       "      <td>1441384.0</td>\n",
       "      <td>106358.0</td>\n",
       "      <td>499900.0</td>\n",
       "      <td>592800.0</td>\n",
       "      <td>329800.0</td>\n",
       "      <td>197183535</td>\n",
       "      <td>37290709</td>\n",
       "      <td>47793785</td>\n",
       "      <td>0.253520</td>\n",
       "      <td>1.589672</td>\n",
       "      <td>0.690048</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2009</th>\n",
       "      <td>1553574.0</td>\n",
       "      <td>187886.0</td>\n",
       "      <td>1365688.0</td>\n",
       "      <td>1448239.0</td>\n",
       "      <td>105335.0</td>\n",
       "      <td>490000.0</td>\n",
       "      <td>584800.0</td>\n",
       "      <td>341200.0</td>\n",
       "      <td>197274549</td>\n",
       "      <td>37656592</td>\n",
       "      <td>49327489</td>\n",
       "      <td>0.248385</td>\n",
       "      <td>1.552982</td>\n",
       "      <td>0.691704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2010</th>\n",
       "      <td>1552669.0</td>\n",
       "      <td>190641.0</td>\n",
       "      <td>1362028.0</td>\n",
       "      <td>1447766.0</td>\n",
       "      <td>104903.0</td>\n",
       "      <td>484400.0</td>\n",
       "      <td>572700.0</td>\n",
       "      <td>345800.0</td>\n",
       "      <td>197394319</td>\n",
       "      <td>38015461</td>\n",
       "      <td>50754069</td>\n",
       "      <td>0.245397</td>\n",
       "      <td>1.506492</td>\n",
       "      <td>0.681325</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011</th>\n",
       "      <td>1538847.0</td>\n",
       "      <td>197050.0</td>\n",
       "      <td>1341797.0</td>\n",
       "      <td>1435141.0</td>\n",
       "      <td>103706.0</td>\n",
       "      <td>474300.0</td>\n",
       "      <td>557100.0</td>\n",
       "      <td>347800.0</td>\n",
       "      <td>197519026</td>\n",
       "      <td>38393758</td>\n",
       "      <td>51906353</td>\n",
       "      <td>0.240129</td>\n",
       "      <td>1.451017</td>\n",
       "      <td>0.670053</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2012</th>\n",
       "      <td>1512430.0</td>\n",
       "      <td>196574.0</td>\n",
       "      <td>1315856.0</td>\n",
       "      <td>1411076.0</td>\n",
       "      <td>101354.0</td>\n",
       "      <td>466600.0</td>\n",
       "      <td>537800.0</td>\n",
       "      <td>340300.0</td>\n",
       "      <td>197701109</td>\n",
       "      <td>38776276</td>\n",
       "      <td>52993496</td>\n",
       "      <td>0.236013</td>\n",
       "      <td>1.386931</td>\n",
       "      <td>0.642154</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2013</th>\n",
       "      <td>1520403.0</td>\n",
       "      <td>195098.0</td>\n",
       "      <td>1325305.0</td>\n",
       "      <td>1416102.0</td>\n",
       "      <td>104301.0</td>\n",
       "      <td>463900.0</td>\n",
       "      <td>529900.0</td>\n",
       "      <td>341200.0</td>\n",
       "      <td>197777454</td>\n",
       "      <td>39135988</td>\n",
       "      <td>54064149</td>\n",
       "      <td>0.234557</td>\n",
       "      <td>1.353997</td>\n",
       "      <td>0.631102</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2014</th>\n",
       "      <td>1507781.0</td>\n",
       "      <td>191374.0</td>\n",
       "      <td>1316407.0</td>\n",
       "      <td>1401685.0</td>\n",
       "      <td>106096.0</td>\n",
       "      <td>461500.0</td>\n",
       "      <td>518700.0</td>\n",
       "      <td>338900.0</td>\n",
       "      <td>197902336</td>\n",
       "      <td>39507913</td>\n",
       "      <td>55189962</td>\n",
       "      <td>0.233196</td>\n",
       "      <td>1.312902</td>\n",
       "      <td>0.614061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015</th>\n",
       "      <td>1476847.0</td>\n",
       "      <td>178688.0</td>\n",
       "      <td>1298159.0</td>\n",
       "      <td>1371879.0</td>\n",
       "      <td>104968.0</td>\n",
       "      <td>450200.0</td>\n",
       "      <td>499400.0</td>\n",
       "      <td>333200.0</td>\n",
       "      <td>197964402</td>\n",
       "      <td>39876758</td>\n",
       "      <td>56338521</td>\n",
       "      <td>0.227415</td>\n",
       "      <td>1.252359</td>\n",
       "      <td>0.591425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2016</th>\n",
       "      <td>1458173.0</td>\n",
       "      <td>171482.0</td>\n",
       "      <td>1286691.0</td>\n",
       "      <td>1352684.0</td>\n",
       "      <td>105489.0</td>\n",
       "      <td>439800.0</td>\n",
       "      <td>486900.0</td>\n",
       "      <td>339300.0</td>\n",
       "      <td>197969608</td>\n",
       "      <td>40229236</td>\n",
       "      <td>57470287</td>\n",
       "      <td>0.222155</td>\n",
       "      <td>1.210314</td>\n",
       "      <td>0.590392</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Total   Federal      State       Male    Female     White     Black  \\\n",
       "Year                                                                            \n",
       "2006  1504598.0  173533.0  1331065.0  1401261.0  103337.0  507100.0  590300.0   \n",
       "2007  1532851.0  179204.0  1353647.0  1427088.0  105763.0  499800.0  592900.0   \n",
       "2008  1547742.0  182333.0  1365409.0  1441384.0  106358.0  499900.0  592800.0   \n",
       "2009  1553574.0  187886.0  1365688.0  1448239.0  105335.0  490000.0  584800.0   \n",
       "2010  1552669.0  190641.0  1362028.0  1447766.0  104903.0  484400.0  572700.0   \n",
       "2011  1538847.0  197050.0  1341797.0  1435141.0  103706.0  474300.0  557100.0   \n",
       "2012  1512430.0  196574.0  1315856.0  1411076.0  101354.0  466600.0  537800.0   \n",
       "2013  1520403.0  195098.0  1325305.0  1416102.0  104301.0  463900.0  529900.0   \n",
       "2014  1507781.0  191374.0  1316407.0  1401685.0  106096.0  461500.0  518700.0   \n",
       "2015  1476847.0  178688.0  1298159.0  1371879.0  104968.0  450200.0  499400.0   \n",
       "2016  1458173.0  171482.0  1286691.0  1352684.0  105489.0  439800.0  486900.0   \n",
       "\n",
       "      Hispanic  white_only_pop  black_only_pop  Hispanic_pop  White_pct  \\\n",
       "Year                                                                      \n",
       "2006  313600.0       196832697        36520961      44606305   0.257630   \n",
       "2007  330400.0       197011394        36905758      46196853   0.253691   \n",
       "2008  329800.0       197183535        37290709      47793785   0.253520   \n",
       "2009  341200.0       197274549        37656592      49327489   0.248385   \n",
       "2010  345800.0       197394319        38015461      50754069   0.245397   \n",
       "2011  347800.0       197519026        38393758      51906353   0.240129   \n",
       "2012  340300.0       197701109        38776276      52993496   0.236013   \n",
       "2013  341200.0       197777454        39135988      54064149   0.234557   \n",
       "2014  338900.0       197902336        39507913      55189962   0.233196   \n",
       "2015  333200.0       197964402        39876758      56338521   0.227415   \n",
       "2016  339300.0       197969608        40229236      57470287   0.222155   \n",
       "\n",
       "      Black_pct  Hispanic_pct  \n",
       "Year                           \n",
       "2006   1.616332      0.703040  \n",
       "2007   1.606524      0.715200  \n",
       "2008   1.589672      0.690048  \n",
       "2009   1.552982      0.691704  \n",
       "2010   1.506492      0.681325  \n",
       "2011   1.451017      0.670053  \n",
       "2012   1.386931      0.642154  \n",
       "2013   1.353997      0.631102  \n",
       "2014   1.312902      0.614061  \n",
       "2015   1.252359      0.591425  \n",
       "2016   1.210314      0.590392  "
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "race_sex_pop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x244fc094d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "with sns.axes_style(\"whitegrid\"):\n",
    "    fig, ax = plt.subplots(figsize=(10,7))\n",
    "    ax.plot(race_sex_pop['White_pct'])\n",
    "    ax.plot(race_sex_pop['Black_pct'])\n",
    "    ax.plot(race_sex_pop['Hispanic_pct'])\n",
    "    ax.set_title('% of Total US Racial Population in Prison',fontsize=14)\n",
    "    ax.set_xlabel('Year', fontsize=14)\n",
    "    ax.set_ylabel('Percent of Racial Population In Prison [%]', fontsize=14)\n",
    "    ax.legend(fontsize=14)\n",
    "    ax.set_ylim([0, 1.1*max([race_sex_pop['White_pct'].max(), \n",
    "                             race_sex_pop['Black_pct'].max(),\n",
    "                             race_sex_pop['Hispanic_pct'].max()])])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Average percentage of total    black population in prison: 1.440 %\n",
      "Average percentage of total Hispanic population in prison: 0.656 %\n",
      "Average percentage of total    white population in prison: 0.241 %\n"
     ]
    }
   ],
   "source": [
    "print('Average percentage of total {:>8s} population in prison: {:0.3f} %'\n",
    "      .format('black', race_sex_pop['Black_pct'].mean()))\n",
    "print('Average percentage of total {:8s} population in prison: {:0.3f} %'\n",
    "      .format('Hispanic', race_sex_pop['Hispanic_pct'].mean()))\n",
    "print('Average percentage of total {:>8s} population in prison: {:0.3f} %'\n",
    "      .format('white', race_sex_pop['White_pct'].mean()))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Change in percentage of total    black population in prison between 2006 to 2016: -0.406 %\n",
      "Change in percentage of total Hispanic population in prison between 2006 to 2016: -0.113 %\n",
      "Change in percentage of total    white population in prison between 2006 to 2016: -0.035 %\n"
     ]
    }
   ],
   "source": [
    "print('Change in percentage of total {:>8s} population in prison between 2006 to 2016: {:0.3f} %'\n",
    "      .format('black', race_sex_pop['Black_pct']['2016'] - race_sex_pop['Black_pct']['2006']))\n",
    "print('Change in percentage of total {:>8s} population in prison between 2006 to 2016: {:0.3f} %'\n",
    "      .format('Hispanic', race_sex_pop['Hispanic_pct']['2016'] - race_sex_pop['Hispanic_pct']['2006']))\n",
    "print('Change in percentage of total {:>8s} population in prison between 2006 to 2016: {:0.3f} %'\n",
    "      .format('white', race_sex_pop['White_pct']['2016'] - race_sex_pop['White_pct']['2006']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "That's an extremely large difference. A randomly selected black person is nearly 6 times more likely to be in prison than a randomly selected white person and more than twice as likely as a randomly selected Hispanic person. \n",
    "\n",
    "That motivates the question: Why is there such a large difference between these populations? \n",
    "\n",
    "To investigate that question, it would be useful to investigate other questions, such as:\n",
    "* Does the criminal justice system treat different racial groups differently?\n",
    "* Is the average quality of education comparable for different racial populations?\n",
    "* Is the distribution of quality of education comparable for different racial populations?\n",
    "* Is the average quality of employment opportunity comparable for different racial populations?\n",
    "* Is the distribution of quality of employment opportunities comparable for different racial populations?\n",
    "\n",
    "To Be Continued."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:py36]",
   "language": "python",
   "name": "conda-env-py36-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}