forOXTWest.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Further analyses for OXT West regarding computer ownership and internet access in Oakland and other cities\n",
    "# http://www.otxwest.org\n",
    "# American Community Survey data from IPUMS: https://usa.ipums.org/usa/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 276,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Populating the interactive namespace from numpy and matplotlib\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "WARNING: pylab import has clobbered these variables: ['test']\n",
      "`%matplotlib` prevents importing * from pylab and numpy\n"
     ]
    }
   ],
   "source": [
    "import csv\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import os\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.patches as mpatches\n",
    "% pylab inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 277,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "mypath = '/Users/laura/Dropbox/posting/OXTWest'\n",
    "path_acs_internet = os.path.join(mypath, 'oxtwest.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 278,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "oxt = pd.read_csv(path_acs_internet)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 279,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "['year', 'datanum', 'serial', 'hhwt', 'city', 'gq', 'rentgrs', 'foodstmp', 'lingisol', 'cilaptop', 'cihand', 'ciothcomp', 'cinethh', 'cimodem', 'cisat', 'cidsl', 'cifiber', 'cibrdbnd', 'cidial', 'ciothsvc', 'multgen', 'multgend', 'pernum', 'perwt', 'relate', 'related', 'age', 'race', 'raced', 'hispan', 'hispand', 'citizen', 'speakeng', 'empstat', 'empstatd', 'poverty']\n",
      "0          Oakland, CA\n",
      "1          Oakland, CA\n",
      "2    San Francisco, CA\n",
      "3    San Francisco, CA\n",
      "4    San Francisco, CA\n",
      "Name: city, dtype: object\n",
      "city\n",
      "Chicago, IL           45243\n",
      "Cleveland, OH          8109\n",
      "Detroit, MI            9561\n",
      "Flint, MI              1362\n",
      "Manchester, NH         1872\n",
      "Miami, FL              7715\n",
      "Minneapolis, MN        5094\n",
      "New Orleans, LA        9575\n",
      "New York, NY         140629\n",
      "Oakland, CA            9093\n",
      "Providence, RI         3168\n",
      "San Francisco, CA     15451\n",
      "Seattle, WA           11424\n",
      "Name: n, dtype: int64\n",
      "268296\n"
     ]
    }
   ],
   "source": [
    "print list(oxt.columns.values) # var names\n",
    "print oxt['city'].head()\n",
    "oxt['n'] = 1\n",
    "print oxt.groupby(by=['city'])['n'].sum() # number of people in the sample each city\n",
    "print len(oxt.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 201,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "# What does access to Internet at home look like for poor Oakland residents based on age?\n",
    "# first creating variables needed"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 280,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cilaptop\n",
      "N/A (GQ)     2.557585\n",
      "No          20.738400\n",
      "Yes         83.519870\n",
      "Name: perwt, dtype: float64\n",
      "   N/A (GQ)  no_internet  Yes, with a subscription to an Internet Service  \\\n",
      "0         0            1                                                0   \n",
      "1         0            1                                                0   \n",
      "2         0            0                                                1   \n",
      "3         0            0                                                1   \n",
      "4         0            0                                                1   \n",
      "\n",
      "   Yes, without a subscription to an Internet Service  \n",
      "0                                                  0   \n",
      "1                                                  0   \n",
      "2                                                  0   \n",
      "3                                                  0   \n",
      "4                                                  0   \n"
     ]
    }
   ],
   "source": [
    "# internet access variable codebook\n",
    "# 0: N/A (GQ)\n",
    "# 1: Yes, with a subscription to an Internet Service\n",
    "# 2: Yes, without a subscription to an Internet Service\n",
    "# 3: No Internet access at this house, apartment, or mobile home\n",
    "\n",
    "print oxt.groupby(by=['cilaptop'])['perwt'].sum().div(oxt.hhwt.sum()) * 100 # response options\n",
    "\n",
    "test = pd.get_dummies(oxt['cinethh'])\n",
    "test.rename(columns={'No Internet access at this house, apartment, or mobile home':'no_internet'}, inplace=True)\n",
    "print test.head()\n",
    "# keeping just the second column, merging to main data set\n",
    "test = test[['no_internet']]\n",
    "oxt = pd.concat([oxt, test], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 281,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "age\n",
      "1      2930\n",
      "10     2958\n",
      "11     2828\n",
      "12     2848\n",
      "13     2827\n",
      "14     2775\n",
      "15     2886\n",
      "16     3001\n",
      "17     3118\n",
      "18     3750\n",
      "19     3744\n",
      "2      2842\n",
      "20     3771\n",
      "21     3699\n",
      "22     4062\n",
      "...\n",
      "82                           1014\n",
      "83                            972\n",
      "84                            955\n",
      "85                            867\n",
      "86                            878\n",
      "87                            724\n",
      "88                            653\n",
      "89                            539\n",
      "9                            2798\n",
      "90 (90+ in 1980 and 1990)     459\n",
      "91                            238\n",
      "93                            103\n",
      "94                            753\n",
      "95                            800\n",
      "Less than 1 year old         2915\n",
      "Name: n, Length: 95, dtype: int64\n",
      "268296\n"
     ]
    }
   ],
   "source": [
    "# Age variable\n",
    "print oxt.groupby(by=['age'])['n'].sum() # absolute/sampled numbers\n",
    "oxt.age[oxt.age == 'Less than 1 year old'] = 0 # recoding\n",
    "oxt.age[oxt.age == '90 (90+ in 1980 and 1990)'] = 91 # recoding\n",
    "oxt.age = oxt.age.astype(int)\n",
    "\n",
    "# Age variable\n",
    "oxt.loc[(oxt['age'] >= 0) & (oxt['age'] <= 10), 'age_ind'] = '0-10'\n",
    "oxt.loc[(oxt['age'] >= 11) & (oxt['age'] <= 14), 'age_ind'] = '11-14'\n",
    "oxt.loc[(oxt['age'] >= 15) & (oxt['age'] <= 18), 'age_ind'] = '15-18'\n",
    "oxt.loc[(oxt['age'] >= 19) & (oxt['age'] <= 22), 'age_ind'] = '19-22'\n",
    "oxt.loc[(oxt['age'] >= 22) & (oxt['age'] <= 25), 'age_ind'] = '22-25'\n",
    "oxt.loc[(oxt['age'] >= 21) & (oxt['age'] <= 29), 'age_ind'] = '26-29'\n",
    "oxt.loc[(oxt['age'] >= 30) & (oxt['age'] <= 39), 'age_ind'] = '30-39'\n",
    "oxt.loc[(oxt['age'] >= 40) & (oxt['age'] <= 49), 'age_ind'] = '40-49'\n",
    "oxt.loc[(oxt['age'] >= 50) & (oxt['age'] <= 59), 'age_ind'] = '50-59'\n",
    "oxt.loc[(oxt['age'] >= 61), 'age_ind'] = '61+'\n",
    "print len(oxt.index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 282,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "268296\n"
     ]
    }
   ],
   "source": [
    "# just Oakland:\n",
    "oxt_oakland = oxt[oxt['city'] == 'Oakland, CA']\n",
    "print len(oxt.index)\n",
    "\n",
    "# just Poor:\n",
    "oxt_oakland_poor = oxt_oakland[oxt_oakland.poverty < 100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 283,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "age_ind\n",
      "0-10       0.317152\n",
      "11-14      0.329046\n",
      "15-18      0.180771\n",
      "19-22      0.155227\n",
      "26-29      0.220203\n",
      "30-39      0.273447\n",
      "40-49      0.264167\n",
      "50-59      0.351251\n",
      "61+        0.478935\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# defining the function to summarize weighted internet access in the home by age group\n",
    "grouped = oxt_oakland_poor.groupby('age_ind')\n",
    "minifun = lambda x: np.average(x.no_internet, weights = x.perwt)\n",
    "print str(grouped.apply(minifun)) # proportion of poor Oaklanders without internet access in their home"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 284,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "relate\n",
      "Child                        69455\n",
      "Child-in-law                  1108\n",
      "Grandchild                    6967\n",
      "Head/Householder            105917\n",
      "Institutional inmates         4123\n",
      "Other non-relatives          13647\n",
      "Other relatives               5187\n",
      "Parent                        4451\n",
      "Parent-in-Law                  771\n",
      "Partner, friend, visitor     13274\n",
      "Sibling                       4500\n",
      "Sibling-in-Law                 749\n",
      "Spouse                       38147\n",
      "Name: n, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# doing the same thing for just households...\n",
    "# data set for households - just keeping the household head as representative (relate == 1)\n",
    "print oxt.groupby(by=['relate'])['n'].sum()\n",
    "oxt_hh_oakland_poor = oxt_oakland_poor[oxt_oakland_poor['relate'] == 'Head/Householder']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "age_ind\n",
      "15-18      1.000000\n",
      "19-22      0.388397\n",
      "26-29      0.237574\n",
      "30-39      0.316932\n",
      "40-49      0.334739\n",
      "50-59      0.502105\n",
      "61+        0.650864\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# defining the function to summarize weighted internet access in the home by age group - unit is household!\n",
    "grouped = oxt_hh_oakland_poor.groupby('age_ind')\n",
    "minifun = lambda x: np.average(x.no_internet, weights = x.hhwt)\n",
    "print str(grouped.apply(minifun)) # proportion of poor Oakland households without internet access in their home"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# How does computer ownership of poor residents in Oakland compare with other US cities?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 286,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "cilaptop_ind\n",
      "0                58770\n",
      "1               209526\n",
      "Name: n, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# creating a NO laptop variable\n",
    "oxt = pd.concat([oxt, pd.get_dummies(oxt[\"cilaptop\"])[[2]]], axis = 1)\n",
    "# renaming\n",
    "oxt = oxt.rename(columns = {'Yes':'cilaptop_ind'})\n",
    "\n",
    "print oxt.groupby(by=['cilaptop_ind'])['n'].sum()\n",
    "\n",
    "oxt.loc[(oxt['cilaptop_ind'] == 1), 'no_comp'] = 0\n",
    "oxt.loc[(oxt['cilaptop_ind'] == 0), 'no_comp'] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 288,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "no_comp\n",
      "0          209526\n",
      "1           58770\n",
      "Name: n, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print oxt.groupby(by = ['no_comp'])['n'].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 289,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "city\n",
      "Chicago, IL           45243\n",
      "Cleveland, OH          8109\n",
      "Detroit, MI            9561\n",
      "Flint, MI              1362\n",
      "Manchester, NH         1872\n",
      "Miami, FL              7715\n",
      "Minneapolis, MN        5094\n",
      "New Orleans, LA        9575\n",
      "New York, NY         140629\n",
      "Oakland, CA            9093\n",
      "Providence, RI         3168\n",
      "San Francisco, CA     15451\n",
      "Seattle, WA           11424\n",
      "Name: n, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "# Summarizing computer ownership among the poor in the 6 cities available\n",
    "print oxt.groupby(by = ['city'])['n'].sum() # absolute/sampled numbers\n",
    "# keeping just poor\n",
    "oxt_poor = oxt[oxt.poverty < 100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 290,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "city\n",
      "Chicago, IL          0.463431\n",
      "Cleveland, OH        0.546172\n",
      "Detroit, MI          0.504711\n",
      "Flint, MI            0.559227\n",
      "Manchester, NH       0.376820\n",
      "Miami, FL            0.470422\n",
      "Minneapolis, MN      0.397932\n",
      "New Orleans, LA      0.577662\n",
      "New York, NY         0.383169\n",
      "Oakland, CA          0.434421\n",
      "Providence, RI       0.555197\n",
      "San Francisco, CA    0.381865\n",
      "Seattle, WA          0.356036\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Function to summarize proportion of poor in the major cities that own computer\n",
    "grouped = oxt_poor.groupby('city')\n",
    "minifun = lambda x: np.average(x.no_comp, weights = x.perwt)\n",
    "print str(grouped.apply(minifun))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 291,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Getting ready to graph\n",
    "comp_graph = pd.DataFrame((grouped.apply(minifun)))\n",
    "comp_graph.rename(columns = {0: 'comp'}, inplace = True)\n",
    "comp_graph['comp'] = comp_graph['comp'] * 100\n",
    "comp_graph = comp_graph.sort(['comp'], ascending = [0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 292,
   "metadata": {
    "collapsed": false,
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbUAAAFwCAYAAAA/j+XaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nDDVJUhmGmiSpDENNklTGFwwjm9ULnmkQAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x11125dad0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = comp_graph['comp'].plot(kind = 'bar', color = ['b', 'b', 'b', 'b', 'b', 'b', 'b', 'r', 'b', 'b', 'b', 'b', 'b'])\n",
    "ax.set_ylabel(\"no computer (%)\")\n",
    "plt.suptitle('Percent of poor residents without a computer by city', size = 16) # INDIVIDUAL LEVEL\n",
    "\n",
    "plt.savefig('/Users/laura/Dropbox/posting/OXTWest/comp_ind.png', bbox_inches = 'tight', dpi = 200)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 293,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# At the household level (\"proportion of poor households with no compuer\")\n",
    "oxt_poor_hh = oxt_poor[oxt_poor['relate'] == 'Head/Householder']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 294,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "city\n",
      "Chicago, IL          0.479076\n",
      "Cleveland, OH        0.575273\n",
      "Detroit, MI          0.578657\n",
      "Flint, MI            0.609571\n",
      "Manchester, NH       0.376173\n",
      "Miami, FL            0.548430\n",
      "Minneapolis, MN      0.366480\n",
      "New Orleans, LA      0.603201\n",
      "New York, NY         0.417508\n",
      "Oakland, CA          0.484716\n",
      "Providence, RI       0.511022\n",
      "San Francisco, CA    0.423388\n",
      "Seattle, WA          0.266391\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "grouped = oxt_poor_hh.groupby('city')\n",
    "minifun = lambda x: np.average(x.no_comp, weights=x.hhwt)\n",
    "print str(grouped.apply(minifun))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                          0\n",
      "city                       \n",
      "Chicago, IL        0.479076\n",
      "Cleveland, OH      0.575273\n",
      "Detroit, MI        0.578657\n",
      "Flint, MI          0.609571\n",
      "Manchester, NH     0.376173\n",
      "Miami, FL          0.548430\n",
      "Minneapolis, MN    0.366480\n",
      "New Orleans, LA    0.603201\n",
      "New York, NY       0.417508\n",
      "Oakland, CA        0.484716\n",
      "Providence, RI     0.511022\n",
      "San Francisco, CA  0.423388\n",
      "Seattle, WA        0.266391\n"
     ]
    }
   ],
   "source": [
    "# getting ready to graph\n",
    "comp_graph_hh = pd.DataFrame((grouped.apply(minifun)))\n",
    "print comp_graph_hh"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 296,
   "metadata": {
    "collapsed": false
   },
   "outputs": [],
   "source": [
    "comp_graph_hh.rename(columns = {0: 'comp'}, inplace = True)\n",
    "comp_graph_hh['comp'] = comp_graph_hh['comp'] * 100\n",
    "comp_graph_hh = comp_graph_hh.sort(['comp'], ascending = [0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 297,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.text.Text at 0x111199210>"
      ]
     },
     "execution_count": 297,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAccAAAFwCAYAAADNM7B/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nVdtAO+JhbgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10d8568d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = comp_graph_hh['comp'].plot(kind = 'bar', color = ['b', 'b', 'b', 'b', 'b', 'b', 'r', 'b', 'b', 'b', 'b', 'b', 'b'])\n",
    "ax.set_ylabel(\"no computer (%)\")\n",
    "plt.suptitle('Percent of poor households without a computer by city', size=16) # household level"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Are there other cities in the East Bay that have data? (Richmond, Pittsburg, Hayward, etc.) NO"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# What does computer ownership look like with the employment status (EMPSTAT) variable\n",
    "# NOTE: I am doing this just for Oakland"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 298,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "empstat\n",
      "Employed              4493\n",
      "Not in labor force    2587\n",
      "Unemployed             443\n",
      "Name: n, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print oxt_oakland.groupby(by = ['empstat'])['n'].sum() # absolute/sampled numbers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 299,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "empstat\n",
      "Employed              0.120885\n",
      "Not in labor force    0.302571\n",
      "Unemployed            0.253451\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Proportion of each empstat category with computer\n",
    "# just Oakland again because no_comp has been created:\n",
    "oxt_oakland = oxt[oxt['city'] == 'Oakland, CA']\n",
    "grouped = oxt_oakland.groupby('empstat')\n",
    "minifun = lambda x: np.average(x.no_comp, weights = x.perwt)\n",
    "print str(grouped.apply(minifun)) # proportion of Oaklanders without internet access in their home"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 300,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "empstat\n",
      "Employed              0.120145\n",
      "Not in labor force    0.386254\n",
      "Unemployed            0.311451\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# at the household level - whether or not the head of household is employed\n",
    "oxt_oakland_hh = oxt_oakland[oxt_oakland['relate'] == 'Head/Householder']\n",
    "grouped = oxt_oakland_hh.groupby('empstat')\n",
    "minifun = lambda x: np.average(x.no_comp, weights = x.hhwt)\n",
    "print str(grouped.apply(minifun))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# What does Internet and computer access look like by race/ethnicity?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 301,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "race\n",
      "American Indian or Alaska Native      1042\n",
      "Black/Negro                          71323\n",
      "Chinese                              15870\n",
      "Japanese                               908\n",
      "Other Asian or Pacific Islander      16633\n",
      "Other race, nec                      23943\n",
      "Three or more major races             1026\n",
      "Two major races                       7824\n",
      "White                               129727\n",
      "Name: n, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print oxt.groupby(by = ['race'])['n'].sum() # absolute/sampled numbers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 302,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "hispan\n",
      "Cuban             3454\n",
      "Mexican          17238\n",
      "Not Hispanic    208901\n",
      "Other            24761\n",
      "Puerto Rican     13942\n",
      "Name: n, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "print oxt.groupby(by = ['hispan'])['n'].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "race_var\n",
      "Black only    1882\n",
      "Hispanic      2118\n",
      "Other         2110\n",
      "White only    2983\n",
      "Name: n, dtype: int64\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/laura/anaconda/lib/python2.7/site-packages/IPython/kernel/__main__.py:5: SettingWithCopyWarning: A value is trying to be set on a copy of a slice from a DataFrame\n"
     ]
    }
   ],
   "source": [
    "# variable for Black/Negro only, white only, Hispanic origin only, other (including mixed race) - only in Oakland\n",
    "oxt_oakland.loc[(oxt_oakland['race'] == \"Black/Negro\"), 'race_var'] = 'Black only'\n",
    "oxt_oakland.loc[(oxt_oakland['race'] == \"White\"), 'race_var'] = 'White only'\n",
    "oxt_oakland.loc[(oxt_oakland['hispan'] != \"Not Hispanic\"), 'race_var'] = 'Hispanic'\n",
    "oxt_oakland['race_var'].fillna('Other', inplace=True)\n",
    "\n",
    "print oxt_oakland.groupby(by = ['race_var'])['n'].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 304,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "race_var\n",
      "Black only    0.319171\n",
      "Hispanic      0.247507\n",
      "Other         0.157748\n",
      "White only    0.072315\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Computer by race/ethnicity\n",
    "grouped = oxt_oakland.groupby('race_var')\n",
    "minifun = lambda x: np.average(x.no_comp, weights = x.perwt)\n",
    "print str(grouped.apply(minifun))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 305,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "race_var\n",
      "Black only    0.232598\n",
      "Hispanic      0.219699\n",
      "Other         0.112911\n",
      "White only    0.049504\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# Internet by race/ethnicity\n",
    "grouped = oxt_oakland.groupby('race_var')\n",
    "minifun = lambda x: np.average(x.no_internet, weights = x.perwt)\n",
    "print str(grouped.apply(minifun))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 306,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "race_var\n",
      "Black only    0.356790\n",
      "Hispanic      0.266187\n",
      "Other         0.203249\n",
      "White only    0.075617\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# At the household level - first computer by race/ethnicity\n",
    "oxt_oakland_hh = oxt_oakland[oxt_oakland['relate'] == 'Head/Householder']\n",
    "grouped = oxt_oakland_hh.groupby('race_var')\n",
    "minifun = lambda x: np.average(x.no_comp, weights = x.hhwt)\n",
    "print str(grouped.apply(minifun))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 307,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "race_var\n",
      "Black only    0.322201\n",
      "Hispanic      0.246084\n",
      "Other         0.158864\n",
      "White only    0.074195\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "# At the household level - then internet by race/ethnicity\n",
    "oxt_oakland_hh = oxt_oakland[oxt_oakland['relate'] == 'Head/Householder']\n",
    "grouped = oxt_oakland.groupby('race_var')\n",
    "minifun = lambda x: np.average(x.no_comp, weights = x.hhwt)\n",
    "print str(grouped.apply(minifun))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 308,
   "metadata": {
    "collapsed": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div style=\"max-height:1000px;max-width:1500px;overflow:auto;\">\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>year</th>\n",
       "      <th>datanum</th>\n",
       "      <th>serial</th>\n",
       "      <th>hhwt</th>\n",
       "      <th>city</th>\n",
       "      <th>gq</th>\n",
       "      <th>rentgrs</th>\n",
       "      <th>foodstmp</th>\n",
       "      <th>lingisol</th>\n",
       "      <th>cilaptop</th>\n",
       "      <th>...</th>\n",
       "      <th>speakeng</th>\n",
       "      <th>empstat</th>\n",
       "      <th>empstatd</th>\n",
       "      <th>poverty</th>\n",
       "      <th>n</th>\n",
       "      <th>no_internet</th>\n",
       "      <th>age_ind</th>\n",
       "      <th>cilaptop_ind</th>\n",
       "      <th>no_comp</th>\n",
       "      <th>race_var</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0 </th>\n",
       "      <td> 2013</td>\n",
       "      <td> 1</td>\n",
       "      <td> 66410</td>\n",
       "      <td> 97</td>\n",
       "      <td> Oakland, CA</td>\n",
       "      <td> Households under 1970 definition</td>\n",
       "      <td> 790</td>\n",
       "      <td> Yes</td>\n",
       "      <td>     Linguistically isolated</td>\n",
       "      <td>  No</td>\n",
       "      <td>...</td>\n",
       "      <td>        Yes, but not well</td>\n",
       "      <td> Not in labor force</td>\n",
       "      <td>   Not in Labor Force</td>\n",
       "      <td>  60</td>\n",
       "      <td> 1</td>\n",
       "      <td> 1</td>\n",
       "      <td>   61+</td>\n",
       "      <td> 0</td>\n",
       "      <td> 1</td>\n",
       "      <td>      Other</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1 </th>\n",
       "      <td> 2013</td>\n",
       "      <td> 1</td>\n",
       "      <td> 66410</td>\n",
       "      <td> 97</td>\n",
       "      <td> Oakland, CA</td>\n",
       "      <td> Households under 1970 definition</td>\n",
       "      <td> 790</td>\n",
       "      <td> Yes</td>\n",
       "      <td>     Linguistically isolated</td>\n",
       "      <td>  No</td>\n",
       "      <td>...</td>\n",
       "      <td>   Does not speak English</td>\n",
       "      <td> Not in labor force</td>\n",
       "      <td>   Not in Labor Force</td>\n",
       "      <td>  60</td>\n",
       "      <td> 1</td>\n",
       "      <td> 1</td>\n",
       "      <td>   61+</td>\n",
       "      <td> 0</td>\n",
       "      <td> 1</td>\n",
       "      <td>      Other</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td> 2013</td>\n",
       "      <td> 1</td>\n",
       "      <td> 66474</td>\n",
       "      <td> 62</td>\n",
       "      <td> Oakland, CA</td>\n",
       "      <td> Households under 1970 definition</td>\n",
       "      <td>   0</td>\n",
       "      <td>  No</td>\n",
       "      <td> Not linguistically isolated</td>\n",
       "      <td> Yes</td>\n",
       "      <td>...</td>\n",
       "      <td> Yes, speaks only English</td>\n",
       "      <td> Not in labor force</td>\n",
       "      <td>   Not in Labor Force</td>\n",
       "      <td> 117</td>\n",
       "      <td> 1</td>\n",
       "      <td> 0</td>\n",
       "      <td>   61+</td>\n",
       "      <td> 1</td>\n",
       "      <td> 0</td>\n",
       "      <td> Black only</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td> 2013</td>\n",
       "      <td> 1</td>\n",
       "      <td> 66558</td>\n",
       "      <td> 55</td>\n",
       "      <td> Oakland, CA</td>\n",
       "      <td> Households under 1970 definition</td>\n",
       "      <td>   0</td>\n",
       "      <td>  No</td>\n",
       "      <td> Not linguistically isolated</td>\n",
       "      <td> Yes</td>\n",
       "      <td>...</td>\n",
       "      <td>         Yes, speaks well</td>\n",
       "      <td> Not in labor force</td>\n",
       "      <td>   Not in Labor Force</td>\n",
       "      <td> 220</td>\n",
       "      <td> 1</td>\n",
       "      <td> 0</td>\n",
       "      <td>   61+</td>\n",
       "      <td> 1</td>\n",
       "      <td> 0</td>\n",
       "      <td>      Other</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td> 2013</td>\n",
       "      <td> 1</td>\n",
       "      <td> 66558</td>\n",
       "      <td> 55</td>\n",
       "      <td> Oakland, CA</td>\n",
       "      <td> Households under 1970 definition</td>\n",
       "      <td>   0</td>\n",
       "      <td>  No</td>\n",
       "      <td> Not linguistically isolated</td>\n",
       "      <td> Yes</td>\n",
       "      <td>...</td>\n",
       "      <td>         Yes, speaks well</td>\n",
       "      <td>           Employed</td>\n",
       "      <td> Has job, not working</td>\n",
       "      <td> 220</td>\n",
       "      <td> 1</td>\n",
       "      <td> 0</td>\n",
       "      <td> 50-59</td>\n",
       "      <td> 1</td>\n",
       "      <td> 0</td>\n",
       "      <td>      Other</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 42 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    year  datanum  serial  hhwt         city  \\\n",
       "0   2013        1   66410    97  Oakland, CA   \n",
       "1   2013        1   66410    97  Oakland, CA   \n",
       "13  2013        1   66474    62  Oakland, CA   \n",
       "18  2013        1   66558    55  Oakland, CA   \n",
       "19  2013        1   66558    55  Oakland, CA   \n",
       "\n",
       "                                  gq  rentgrs foodstmp  \\\n",
       "0   Households under 1970 definition      790      Yes   \n",
       "1   Households under 1970 definition      790      Yes   \n",
       "13  Households under 1970 definition        0       No   \n",
       "18  Households under 1970 definition        0       No   \n",
       "19  Households under 1970 definition        0       No   \n",
       "\n",
       "                       lingisol cilaptop   ...                     speakeng  \\\n",
       "0       Linguistically isolated       No   ...            Yes, but not well   \n",
       "1       Linguistically isolated       No   ...       Does not speak English   \n",
       "13  Not linguistically isolated      Yes   ...     Yes, speaks only English   \n",
       "18  Not linguistically isolated      Yes   ...             Yes, speaks well   \n",
       "19  Not linguistically isolated      Yes   ...             Yes, speaks well   \n",
       "\n",
       "               empstat              empstatd poverty  n no_internet age_ind  \\\n",
       "0   Not in labor force    Not in Labor Force      60  1           1     61+   \n",
       "1   Not in labor force    Not in Labor Force      60  1           1     61+   \n",
       "13  Not in labor force    Not in Labor Force     117  1           0     61+   \n",
       "18  Not in labor force    Not in Labor Force     220  1           0     61+   \n",
       "19            Employed  Has job, not working     220  1           0   50-59   \n",
       "\n",
       "   cilaptop_ind no_comp    race_var  \n",
       "0             0       1       Other  \n",
       "1             0       1       Other  \n",
       "13            1       0  Black only  \n",
       "18            1       0       Other  \n",
       "19            1       0       Other  \n",
       "\n",
       "[5 rows x 42 columns]"
      ]
     },
     "execution_count": 308,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(oxt_oakland.index)\n",
    "oxt_oakland.head()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}