Tables.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import modules and set options\n",
    "%matplotlib inline\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import seaborn as sns\n",
    "# import pandas_profiling\n",
    "\n",
    "sns.set(context='notebook', style='ticks')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "analysis_subset = (pd.read_csv('../data/clean/analysis_subset.csv', low_memory=False)\n",
    "                  .rename({'onset_1':'identify_mo'}, axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(33128, 54)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "analysis_subset.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "age_mask = (analysis_subset.age_test>=48) & (analysis_subset.age_test<60)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1424, 54)"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "analysis_subset[age_mask].drop_duplicates(subset='student_idx').shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_4yo = analysis_subset[age_mask].drop_duplicates(subset='student_idx').copy()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Demographics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Male      761\n",
       "Female    663\n",
       "Name: male, dtype: int64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_4yo.male.replace({1: 'Male', 0: 'Female', np.nan: 'Missing'}).value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "White       918\n",
       "Black       150\n",
       "Hispanic    144\n",
       "Other       137\n",
       "Asian        75\n",
       "Name: race, dtype: int64"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_4yo.race.replace({0:'White', 1:'Black', 2:'Hispanic', 3:'Asian', 4:'Other', np.nan:'Missing'}).value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<36 weeks     1060\n",
       "Unknown        228\n",
       ">=36 weeks     136\n",
       "Name: premature, dtype: int64"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_4yo.premature.replace({True: '>=36 weeks', False: '<36 weeks', np.nan: 'Unknown'}).value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1          337\n",
       "2          590\n",
       "3          272\n",
       "4+         160\n",
       "Missing     65\n",
       "Name: sib, dtype: int64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_4yo.sib.replace({0:'1', 1:'2', 2:'3', 3:'4+', np.nan:'Missing'}).value_counts().sort_index()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ideal Participation      455\n",
       "Good Participation       381\n",
       "Average Participation    320\n",
       "Missing                  151\n",
       "Below Average             95\n",
       "Limited Participation     22\n",
       "Name: family_inv, dtype: int64"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "family_lookup = {0:\"Ideal Participation\",\n",
    "1:\"Good Participation\",\n",
    "2:\"Average Participation\",\n",
    "3:\"Below Average\",\n",
    "4:\"Limited Participation\",\n",
    "np.nan:\"Missing\"}\n",
    "\n",
    "data_4yo.family_inv.replace(family_lookup).value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hearing loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "187"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_4yo.deg_hl_below6.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "hl_data = data_4yo.set_index('student_idx')[['bilateral_snhl',\n",
    "                     'bilateral_ansd',\n",
    "                     'bilateral_mixed',\n",
    "                     'bilateral_cond',\n",
    "                     'bilateral_normal',\n",
    "                     'bilateral_unk',\n",
    "                     'unilateral_snhl',\n",
    "                     'unilateral_ansd',\n",
    "                     'unilateral_mixed',\n",
    "                     'unilateral_cond',\n",
    "                     'unilateral_unk',\n",
    "                     'assymetrical']]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "One or both parents with hearing loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    1164\n",
       "1.0      88\n",
       "Name: one_or_both_parent_hl, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_4yo.one_or_both_parent_hl.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "172"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_4yo.one_or_both_parent_hl.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Individuals with no hearing loss type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bilateral_snhl      1089\n",
       "bilateral_cond        43\n",
       "bilateral_ansd        38\n",
       "assymetrical          34\n",
       "bilateral_mixed       27\n",
       "unilateral_unk         0\n",
       "unilateral_cond        0\n",
       "unilateral_mixed       0\n",
       "unilateral_ansd        0\n",
       "unilateral_snhl        0\n",
       "bilateral_unk          0\n",
       "bilateral_normal       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hl_data.sum().astype(int).sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bilateral_snhl      0.76\n",
       "bilateral_cond      0.03\n",
       "bilateral_ansd      0.03\n",
       "assymetrical        0.02\n",
       "bilateral_mixed     0.02\n",
       "unilateral_unk      0.00\n",
       "unilateral_cond     0.00\n",
       "unilateral_mixed    0.00\n",
       "unilateral_ansd     0.00\n",
       "unilateral_snhl     0.00\n",
       "bilateral_unk       0.00\n",
       "bilateral_normal    0.00\n",
       "dtype: float64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hl_data.mean().round(2).sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "tech_data = data_4yo.set_index('student_idx')[['bilateral_ha',\n",
    "                     'bilateral_ci',\n",
    "                     'bimodal',\n",
    "                     'bilateral_other',\n",
    "                     'unilateral_ha',\n",
    "                     'unilateral_ci',\n",
    "                     'unilateral_other']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bilateral_ci        1054\n",
       "bilateral_ha         268\n",
       "bimodal               75\n",
       "unilateral_ci         13\n",
       "unilateral_ha          5\n",
       "unilateral_other       1\n",
       "bilateral_other        1\n",
       "dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tech_data.sum().astype(int).sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bilateral_ci        0.74\n",
       "bilateral_ha        0.19\n",
       "bimodal             0.05\n",
       "unilateral_ci       0.01\n",
       "unilateral_other    0.00\n",
       "unilateral_ha       0.00\n",
       "bilateral_other     0.00\n",
       "dtype: float64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tech_data.mean().round(2).sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Individuals with no technology type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([3517, 3178, 2701, 6132,  296,  139, 3953])"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tech_data[tech_data.sum(1)==0].index.values"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hearing Loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['bilateral_hl_normal', 'unilateral_hl_severe', 'unilateral_hl_mild',\n",
       "       'unilateral_hl_slight', 'bilateral_hl_severe', 'bilateral_hl_slight',\n",
       "       'unilateral_hl_moderate', 'bilateral_hl_mild', 'bilateral_hl_moderate',\n",
       "       'unilateral_hl_modsev', 'bilateral_hl_profound', 'deg_hl_below6',\n",
       "       'unilateral_hl_profound', 'bilateral_hl_modsev'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_4yo.columns[data_4yo.columns.str.contains('_hl_')]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "hl_data = data_4yo.set_index('student_idx')[['bilateral_hl_normal', 'bilateral_hl_slight', 'bilateral_hl_mild',\n",
    "       'bilateral_hl_moderate', 'bilateral_hl_modsev', 'bilateral_hl_severe',\n",
    "       'bilateral_hl_profound', 'unilateral_hl_slight', 'unilateral_hl_mild',\n",
    "       'unilateral_hl_moderate', 'unilateral_hl_modsev',\n",
    "       'unilateral_hl_severe', 'unilateral_hl_profound']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bilateral_hl_profound     487\n",
       "bilateral_hl_moderate     144\n",
       "bilateral_hl_modsev       119\n",
       "bilateral_hl_severe        88\n",
       "bilateral_hl_mild          53\n",
       "bilateral_hl_slight        10\n",
       "unilateral_hl_slight        2\n",
       "unilateral_hl_modsev        1\n",
       "unilateral_hl_mild          1\n",
       "unilateral_hl_profound      0\n",
       "unilateral_hl_severe        0\n",
       "unilateral_hl_moderate      0\n",
       "bilateral_hl_normal         0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hl_data.sum().astype(int).sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bilateral_hl_profound     0.34\n",
       "bilateral_hl_moderate     0.10\n",
       "bilateral_hl_modsev       0.08\n",
       "bilateral_hl_severe       0.06\n",
       "bilateral_hl_mild         0.04\n",
       "bilateral_hl_slight       0.01\n",
       "unilateral_hl_profound    0.00\n",
       "unilateral_hl_severe      0.00\n",
       "unilateral_hl_modsev      0.00\n",
       "unilateral_hl_moderate    0.00\n",
       "unilateral_hl_mild        0.00\n",
       "unilateral_hl_slight      0.00\n",
       "bilateral_hl_normal       0.00\n",
       "dtype: float64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "hl_data.mean().round(2).sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Summary of scores by domain"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_scores = analysis_subset[age_mask].copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_scores.loc[(test_scores.domain=='Language') & (test_scores.test_type=='receptive'), \n",
    "           'domain'] = 'Receptive Language'\n",
    "test_scores.loc[(test_scores.domain=='Language') & (test_scores.test_type=='expressive'), \n",
    "           'domain'] = 'Expressive Language'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>domain</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",
       "      <td>Articulation</td>\n",
       "      <td>927.0</td>\n",
       "      <td>86.4</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>74.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>102.0</td>\n",
       "      <td>123.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Expressive Language</td>\n",
       "      <td>1183.0</td>\n",
       "      <td>86.4</td>\n",
       "      <td>19.8</td>\n",
       "      <td>45.0</td>\n",
       "      <td>71.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>101.0</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Expressive Vocabulary</td>\n",
       "      <td>1275.0</td>\n",
       "      <td>98.2</td>\n",
       "      <td>20.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>113.0</td>\n",
       "      <td>147.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Language</td>\n",
       "      <td>1218.0</td>\n",
       "      <td>88.3</td>\n",
       "      <td>20.5</td>\n",
       "      <td>40.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>88.0</td>\n",
       "      <td>102.8</td>\n",
       "      <td>163.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Receptive Language</td>\n",
       "      <td>1187.0</td>\n",
       "      <td>90.9</td>\n",
       "      <td>19.1</td>\n",
       "      <td>50.0</td>\n",
       "      <td>77.0</td>\n",
       "      <td>92.0</td>\n",
       "      <td>105.0</td>\n",
       "      <td>149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>Receptive Vocabulary</td>\n",
       "      <td>1289.0</td>\n",
       "      <td>95.4</td>\n",
       "      <td>18.7</td>\n",
       "      <td>0.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>97.0</td>\n",
       "      <td>109.0</td>\n",
       "      <td>159.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        count  mean   std   min   25%   50%    75%    max\n",
       "domain                                                                   \n",
       "Articulation            927.0  86.4  20.0   0.0  74.0  89.0  102.0  123.0\n",
       "Expressive Language    1183.0  86.4  19.8  45.0  71.0  86.0  101.0  141.0\n",
       "Expressive Vocabulary  1275.0  98.2  20.0   0.0  86.0  99.0  113.0  147.0\n",
       "Language               1218.0  88.3  20.5  40.0  73.0  88.0  102.8  163.0\n",
       "Receptive Language     1187.0  90.9  19.1  50.0  77.0  92.0  105.0  149.0\n",
       "Receptive Vocabulary   1289.0  95.4  18.7   0.0  83.0  97.0  109.0  159.0"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_scores.groupby('domain').score.describe().round(1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Expressive Vocabulary', 'Receptive Language',\n",
       "       'Expressive Language', 'Language', 'Receptive Vocabulary',\n",
       "       'Articulation'], dtype=object)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_scores.domain.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_scores['domain_labels'] = test_scores.domain.replace({'Expressive Vocabulary': 'Expressive\\nVocabulary', \n",
    "                                                           'Receptive Vocabulary': 'Receptive\\nVocabulary',\n",
    "                                                           'Receptive Language': 'Receptive\\nLanguage', \n",
    "                                                           'Expressive Language': 'Expressive\\nLanguage', \n",
    "                                                           'Language': 'Total\\nLanguage'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(\"domain_labels\", y=\"score\", data=test_scores)\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(12, 5)\n",
    "fig.axes[0].set_xlabel('');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(\"domain_labels\", y=\"score\", data=test_scores, kind=\"box\", color='white', fliersize=0)\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(12, 5)\n",
    "fig.axes[0].set_xlabel('');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Domain-specific summaries\n",
    "\n",
    "Age of amplification greater than 6mo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>domain</th>\n",
       "      <th>age_amp_greater_6</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",
       "      <td rowspan=\"2\" valign=\"top\">Articulation</td>\n",
       "      <td>False</td>\n",
       "      <td>512.0</td>\n",
       "      <td>91.417969</td>\n",
       "      <td>17.980450</td>\n",
       "      <td>40.0</td>\n",
       "      <td>81.00</td>\n",
       "      <td>94.0</td>\n",
       "      <td>106.00</td>\n",
       "      <td>123.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>415.0</td>\n",
       "      <td>80.306024</td>\n",
       "      <td>20.580121</td>\n",
       "      <td>0.0</td>\n",
       "      <td>67.00</td>\n",
       "      <td>83.0</td>\n",
       "      <td>97.00</td>\n",
       "      <td>118.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Expressive Language</td>\n",
       "      <td>False</td>\n",
       "      <td>576.0</td>\n",
       "      <td>90.942708</td>\n",
       "      <td>18.589811</td>\n",
       "      <td>45.0</td>\n",
       "      <td>77.00</td>\n",
       "      <td>92.0</td>\n",
       "      <td>104.00</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>607.0</td>\n",
       "      <td>82.069193</td>\n",
       "      <td>19.899960</td>\n",
       "      <td>45.0</td>\n",
       "      <td>67.00</td>\n",
       "      <td>80.0</td>\n",
       "      <td>96.50</td>\n",
       "      <td>136.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Expressive Vocabulary</td>\n",
       "      <td>False</td>\n",
       "      <td>705.0</td>\n",
       "      <td>101.946099</td>\n",
       "      <td>18.199102</td>\n",
       "      <td>41.0</td>\n",
       "      <td>90.00</td>\n",
       "      <td>103.0</td>\n",
       "      <td>116.00</td>\n",
       "      <td>146.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>570.0</td>\n",
       "      <td>93.471930</td>\n",
       "      <td>21.055957</td>\n",
       "      <td>0.0</td>\n",
       "      <td>80.00</td>\n",
       "      <td>94.0</td>\n",
       "      <td>107.00</td>\n",
       "      <td>147.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Language</td>\n",
       "      <td>False</td>\n",
       "      <td>591.0</td>\n",
       "      <td>92.710660</td>\n",
       "      <td>18.952730</td>\n",
       "      <td>45.0</td>\n",
       "      <td>79.00</td>\n",
       "      <td>94.0</td>\n",
       "      <td>106.00</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>627.0</td>\n",
       "      <td>84.177033</td>\n",
       "      <td>21.051107</td>\n",
       "      <td>40.0</td>\n",
       "      <td>69.00</td>\n",
       "      <td>83.0</td>\n",
       "      <td>98.00</td>\n",
       "      <td>163.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Receptive Language</td>\n",
       "      <td>False</td>\n",
       "      <td>582.0</td>\n",
       "      <td>95.082474</td>\n",
       "      <td>17.413274</td>\n",
       "      <td>50.0</td>\n",
       "      <td>85.00</td>\n",
       "      <td>96.0</td>\n",
       "      <td>107.00</td>\n",
       "      <td>145.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>605.0</td>\n",
       "      <td>86.874380</td>\n",
       "      <td>19.875375</td>\n",
       "      <td>50.0</td>\n",
       "      <td>71.00</td>\n",
       "      <td>88.0</td>\n",
       "      <td>101.00</td>\n",
       "      <td>149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Receptive Vocabulary</td>\n",
       "      <td>False</td>\n",
       "      <td>709.0</td>\n",
       "      <td>99.471086</td>\n",
       "      <td>17.448417</td>\n",
       "      <td>38.0</td>\n",
       "      <td>88.00</td>\n",
       "      <td>100.0</td>\n",
       "      <td>112.00</td>\n",
       "      <td>149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>580.0</td>\n",
       "      <td>90.386207</td>\n",
       "      <td>18.869312</td>\n",
       "      <td>0.0</td>\n",
       "      <td>78.75</td>\n",
       "      <td>91.0</td>\n",
       "      <td>103.25</td>\n",
       "      <td>159.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         count        mean        std   min  \\\n",
       "domain                age_amp_greater_6                                       \n",
       "Articulation          False              512.0   91.417969  17.980450  40.0   \n",
       "                      True               415.0   80.306024  20.580121   0.0   \n",
       "Expressive Language   False              576.0   90.942708  18.589811  45.0   \n",
       "                      True               607.0   82.069193  19.899960  45.0   \n",
       "Expressive Vocabulary False              705.0  101.946099  18.199102  41.0   \n",
       "                      True               570.0   93.471930  21.055957   0.0   \n",
       "Language              False              591.0   92.710660  18.952730  45.0   \n",
       "                      True               627.0   84.177033  21.051107  40.0   \n",
       "Receptive Language    False              582.0   95.082474  17.413274  50.0   \n",
       "                      True               605.0   86.874380  19.875375  50.0   \n",
       "Receptive Vocabulary  False              709.0   99.471086  17.448417  38.0   \n",
       "                      True               580.0   90.386207  18.869312   0.0   \n",
       "\n",
       "                                           25%    50%     75%    max  \n",
       "domain                age_amp_greater_6                               \n",
       "Articulation          False              81.00   94.0  106.00  123.0  \n",
       "                      True               67.00   83.0   97.00  118.0  \n",
       "Expressive Language   False              77.00   92.0  104.00  141.0  \n",
       "                      True               67.00   80.0   96.50  136.0  \n",
       "Expressive Vocabulary False              90.00  103.0  116.00  146.0  \n",
       "                      True               80.00   94.0  107.00  147.0  \n",
       "Language              False              79.00   94.0  106.00  141.0  \n",
       "                      True               69.00   83.0   98.00  163.0  \n",
       "Receptive Language    False              85.00   96.0  107.00  145.0  \n",
       "                      True               71.00   88.0  101.00  149.0  \n",
       "Receptive Vocabulary  False              88.00  100.0  112.00  149.0  \n",
       "                      True               78.75   91.0  103.25  159.0  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(test_scores.assign(age_amp_greater_6=test_scores.age_amp>6)\n",
    "         .groupby(['domain', 'age_amp_greater_6'])\n",
    "         .score.describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(\"domain_labels\", y=\"score\", hue='age_amp_greater_6', data=test_scores.assign(age_amp_greater_6=test_scores.age_amp>6),\n",
    "            kind=\"box\", color='white', fliersize=0)\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(12, 5)\n",
    "fig.axes[0].set_xlabel('');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Age of enrollment greater than 6mo."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "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></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>domain</th>\n",
       "      <th>age_enroll_greater_6</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",
       "      <td rowspan=\"2\" valign=\"top\">Articulation</td>\n",
       "      <td>False</td>\n",
       "      <td>512.0</td>\n",
       "      <td>89.833984</td>\n",
       "      <td>19.649019</td>\n",
       "      <td>0.0</td>\n",
       "      <td>79.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>105.0</td>\n",
       "      <td>123.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>415.0</td>\n",
       "      <td>82.260241</td>\n",
       "      <td>19.561367</td>\n",
       "      <td>40.0</td>\n",
       "      <td>69.0</td>\n",
       "      <td>84.0</td>\n",
       "      <td>97.5</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Expressive Language</td>\n",
       "      <td>False</td>\n",
       "      <td>576.0</td>\n",
       "      <td>89.508681</td>\n",
       "      <td>19.403091</td>\n",
       "      <td>45.0</td>\n",
       "      <td>75.0</td>\n",
       "      <td>91.0</td>\n",
       "      <td>104.0</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>607.0</td>\n",
       "      <td>83.429984</td>\n",
       "      <td>19.674110</td>\n",
       "      <td>45.0</td>\n",
       "      <td>69.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>98.0</td>\n",
       "      <td>136.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Expressive Vocabulary</td>\n",
       "      <td>False</td>\n",
       "      <td>696.0</td>\n",
       "      <td>101.225575</td>\n",
       "      <td>19.908481</td>\n",
       "      <td>0.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>103.0</td>\n",
       "      <td>116.0</td>\n",
       "      <td>146.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>579.0</td>\n",
       "      <td>94.469775</td>\n",
       "      <td>19.428879</td>\n",
       "      <td>0.0</td>\n",
       "      <td>82.0</td>\n",
       "      <td>96.0</td>\n",
       "      <td>107.0</td>\n",
       "      <td>147.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Language</td>\n",
       "      <td>False</td>\n",
       "      <td>591.0</td>\n",
       "      <td>91.294416</td>\n",
       "      <td>19.789347</td>\n",
       "      <td>45.0</td>\n",
       "      <td>77.5</td>\n",
       "      <td>92.0</td>\n",
       "      <td>106.0</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>627.0</td>\n",
       "      <td>85.511962</td>\n",
       "      <td>20.778409</td>\n",
       "      <td>40.0</td>\n",
       "      <td>69.0</td>\n",
       "      <td>85.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>163.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Receptive Language</td>\n",
       "      <td>False</td>\n",
       "      <td>581.0</td>\n",
       "      <td>94.008606</td>\n",
       "      <td>18.401881</td>\n",
       "      <td>50.0</td>\n",
       "      <td>83.0</td>\n",
       "      <td>94.0</td>\n",
       "      <td>107.0</td>\n",
       "      <td>145.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>606.0</td>\n",
       "      <td>87.917492</td>\n",
       "      <td>19.385002</td>\n",
       "      <td>50.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>88.5</td>\n",
       "      <td>101.0</td>\n",
       "      <td>149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Receptive Vocabulary</td>\n",
       "      <td>False</td>\n",
       "      <td>703.0</td>\n",
       "      <td>97.987198</td>\n",
       "      <td>18.882938</td>\n",
       "      <td>0.0</td>\n",
       "      <td>86.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>111.0</td>\n",
       "      <td>149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>True</td>\n",
       "      <td>586.0</td>\n",
       "      <td>92.259386</td>\n",
       "      <td>17.890387</td>\n",
       "      <td>40.0</td>\n",
       "      <td>80.0</td>\n",
       "      <td>93.0</td>\n",
       "      <td>105.0</td>\n",
       "      <td>159.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                            count        mean        std  \\\n",
       "domain                age_enroll_greater_6                                 \n",
       "Articulation          False                 512.0   89.833984  19.649019   \n",
       "                      True                  415.0   82.260241  19.561367   \n",
       "Expressive Language   False                 576.0   89.508681  19.403091   \n",
       "                      True                  607.0   83.429984  19.674110   \n",
       "Expressive Vocabulary False                 696.0  101.225575  19.908481   \n",
       "                      True                  579.0   94.469775  19.428879   \n",
       "Language              False                 591.0   91.294416  19.789347   \n",
       "                      True                  627.0   85.511962  20.778409   \n",
       "Receptive Language    False                 581.0   94.008606  18.401881   \n",
       "                      True                  606.0   87.917492  19.385002   \n",
       "Receptive Vocabulary  False                 703.0   97.987198  18.882938   \n",
       "                      True                  586.0   92.259386  17.890387   \n",
       "\n",
       "                                             min   25%    50%    75%    max  \n",
       "domain                age_enroll_greater_6                                   \n",
       "Articulation          False                  0.0  79.0   93.0  105.0  123.0  \n",
       "                      True                  40.0  69.0   84.0   97.5  121.0  \n",
       "Expressive Language   False                 45.0  75.0   91.0  104.0  141.0  \n",
       "                      True                  45.0  69.0   83.0   98.0  136.0  \n",
       "Expressive Vocabulary False                  0.0  89.0  103.0  116.0  146.0  \n",
       "                      True                   0.0  82.0   96.0  107.0  147.0  \n",
       "Language              False                 45.0  77.5   92.0  106.0  141.0  \n",
       "                      True                  40.0  69.0   85.0  100.0  163.0  \n",
       "Receptive Language    False                 50.0  83.0   94.0  107.0  145.0  \n",
       "                      True                  50.0  73.0   88.5  101.0  149.0  \n",
       "Receptive Vocabulary  False                  0.0  86.0   99.0  111.0  149.0  \n",
       "                      True                  40.0  80.0   93.0  105.0  159.0  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(test_scores.assign(age_enroll_greater_6=test_scores.age_int>6)\n",
    "         .groupby(['domain', 'age_enroll_greater_6'])\n",
    "         .score.describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(\"domain_labels\", y=\"score\", hue='age_enroll_greater_6', data=test_scores.assign(age_enroll_greater_6=test_scores.age_int>6),\n",
    "            kind=\"box\", color='white', fliersize=0)\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(12, 5)\n",
    "fig.axes[0].set_xlabel('');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "(Non-)Profound hearing loss"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "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></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>domain</th>\n",
       "      <th>non_profound_hl</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",
       "      <td rowspan=\"2\" valign=\"top\">Articulation</td>\n",
       "      <td>0.0</td>\n",
       "      <td>380.0</td>\n",
       "      <td>82.931579</td>\n",
       "      <td>20.736583</td>\n",
       "      <td>0.0</td>\n",
       "      <td>68.75</td>\n",
       "      <td>86.0</td>\n",
       "      <td>98.25</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>373.0</td>\n",
       "      <td>86.662198</td>\n",
       "      <td>18.767496</td>\n",
       "      <td>41.0</td>\n",
       "      <td>74.00</td>\n",
       "      <td>88.0</td>\n",
       "      <td>102.00</td>\n",
       "      <td>120.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Expressive Language</td>\n",
       "      <td>0.0</td>\n",
       "      <td>589.0</td>\n",
       "      <td>81.551783</td>\n",
       "      <td>19.493749</td>\n",
       "      <td>45.0</td>\n",
       "      <td>67.00</td>\n",
       "      <td>80.0</td>\n",
       "      <td>96.00</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>545.0</td>\n",
       "      <td>91.297248</td>\n",
       "      <td>18.914759</td>\n",
       "      <td>45.0</td>\n",
       "      <td>77.00</td>\n",
       "      <td>92.0</td>\n",
       "      <td>105.00</td>\n",
       "      <td>136.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Expressive Vocabulary</td>\n",
       "      <td>0.0</td>\n",
       "      <td>552.0</td>\n",
       "      <td>93.862319</td>\n",
       "      <td>20.913248</td>\n",
       "      <td>0.0</td>\n",
       "      <td>80.00</td>\n",
       "      <td>95.5</td>\n",
       "      <td>108.00</td>\n",
       "      <td>146.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>528.0</td>\n",
       "      <td>99.613636</td>\n",
       "      <td>18.193623</td>\n",
       "      <td>0.0</td>\n",
       "      <td>89.00</td>\n",
       "      <td>100.0</td>\n",
       "      <td>113.00</td>\n",
       "      <td>147.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Language</td>\n",
       "      <td>0.0</td>\n",
       "      <td>606.0</td>\n",
       "      <td>83.313531</td>\n",
       "      <td>20.273764</td>\n",
       "      <td>48.0</td>\n",
       "      <td>68.00</td>\n",
       "      <td>83.0</td>\n",
       "      <td>98.00</td>\n",
       "      <td>139.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>563.0</td>\n",
       "      <td>93.349911</td>\n",
       "      <td>19.599008</td>\n",
       "      <td>40.0</td>\n",
       "      <td>80.00</td>\n",
       "      <td>94.0</td>\n",
       "      <td>108.00</td>\n",
       "      <td>163.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Receptive Language</td>\n",
       "      <td>0.0</td>\n",
       "      <td>589.0</td>\n",
       "      <td>86.444822</td>\n",
       "      <td>19.595113</td>\n",
       "      <td>50.0</td>\n",
       "      <td>71.00</td>\n",
       "      <td>88.0</td>\n",
       "      <td>101.00</td>\n",
       "      <td>145.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>549.0</td>\n",
       "      <td>95.320583</td>\n",
       "      <td>17.663190</td>\n",
       "      <td>50.0</td>\n",
       "      <td>83.00</td>\n",
       "      <td>96.0</td>\n",
       "      <td>107.00</td>\n",
       "      <td>149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Receptive Vocabulary</td>\n",
       "      <td>0.0</td>\n",
       "      <td>563.0</td>\n",
       "      <td>90.438721</td>\n",
       "      <td>18.630545</td>\n",
       "      <td>0.0</td>\n",
       "      <td>79.00</td>\n",
       "      <td>91.0</td>\n",
       "      <td>104.00</td>\n",
       "      <td>133.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>529.0</td>\n",
       "      <td>96.879017</td>\n",
       "      <td>16.866244</td>\n",
       "      <td>44.0</td>\n",
       "      <td>86.00</td>\n",
       "      <td>98.0</td>\n",
       "      <td>109.00</td>\n",
       "      <td>159.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                       count       mean        std   min  \\\n",
       "domain                non_profound_hl                                      \n",
       "Articulation          0.0              380.0  82.931579  20.736583   0.0   \n",
       "                      1.0              373.0  86.662198  18.767496  41.0   \n",
       "Expressive Language   0.0              589.0  81.551783  19.493749  45.0   \n",
       "                      1.0              545.0  91.297248  18.914759  45.0   \n",
       "Expressive Vocabulary 0.0              552.0  93.862319  20.913248   0.0   \n",
       "                      1.0              528.0  99.613636  18.193623   0.0   \n",
       "Language              0.0              606.0  83.313531  20.273764  48.0   \n",
       "                      1.0              563.0  93.349911  19.599008  40.0   \n",
       "Receptive Language    0.0              589.0  86.444822  19.595113  50.0   \n",
       "                      1.0              549.0  95.320583  17.663190  50.0   \n",
       "Receptive Vocabulary  0.0              563.0  90.438721  18.630545   0.0   \n",
       "                      1.0              529.0  96.879017  16.866244  44.0   \n",
       "\n",
       "                                         25%    50%     75%    max  \n",
       "domain                non_profound_hl                               \n",
       "Articulation          0.0              68.75   86.0   98.25  121.0  \n",
       "                      1.0              74.00   88.0  102.00  120.0  \n",
       "Expressive Language   0.0              67.00   80.0   96.00  141.0  \n",
       "                      1.0              77.00   92.0  105.00  136.0  \n",
       "Expressive Vocabulary 0.0              80.00   95.5  108.00  146.0  \n",
       "                      1.0              89.00  100.0  113.00  147.0  \n",
       "Language              0.0              68.00   83.0   98.00  139.0  \n",
       "                      1.0              80.00   94.0  108.00  163.0  \n",
       "Receptive Language    0.0              71.00   88.0  101.00  145.0  \n",
       "                      1.0              83.00   96.0  107.00  149.0  \n",
       "Receptive Vocabulary  0.0              79.00   91.0  104.00  133.0  \n",
       "                      1.0              86.00   98.0  109.00  159.0  "
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(test_scores.assign(non_profound_hl=test_scores.deg_hl_below6)\n",
    "         .groupby(['domain', 'non_profound_hl'])\n",
    "         .score.describe())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(\"domain_labels\", y=\"score\", hue='non_profound_hl', data=test_scores.assign(non_profound_hl=test_scores.deg_hl_below6),\n",
    "            kind=\"box\", color='white', fliersize=0)\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(12, 5)\n",
    "fig.axes[0].set_xlabel('');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Mother with(out) college education"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "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></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>domain</th>\n",
       "      <th>mother_college</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",
       "      <td rowspan=\"2\" valign=\"top\">Articulation</td>\n",
       "      <td>0.0</td>\n",
       "      <td>485.0</td>\n",
       "      <td>84.853608</td>\n",
       "      <td>19.612918</td>\n",
       "      <td>0.0</td>\n",
       "      <td>72.00</td>\n",
       "      <td>86.0</td>\n",
       "      <td>101.00</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>130.0</td>\n",
       "      <td>87.723077</td>\n",
       "      <td>19.872468</td>\n",
       "      <td>40.0</td>\n",
       "      <td>75.25</td>\n",
       "      <td>89.5</td>\n",
       "      <td>103.00</td>\n",
       "      <td>121.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Expressive Language</td>\n",
       "      <td>0.0</td>\n",
       "      <td>710.0</td>\n",
       "      <td>84.794366</td>\n",
       "      <td>18.938173</td>\n",
       "      <td>45.0</td>\n",
       "      <td>69.00</td>\n",
       "      <td>85.0</td>\n",
       "      <td>99.75</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>230.0</td>\n",
       "      <td>96.095652</td>\n",
       "      <td>20.233559</td>\n",
       "      <td>45.0</td>\n",
       "      <td>81.00</td>\n",
       "      <td>98.0</td>\n",
       "      <td>111.00</td>\n",
       "      <td>140.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Expressive Vocabulary</td>\n",
       "      <td>0.0</td>\n",
       "      <td>702.0</td>\n",
       "      <td>96.188034</td>\n",
       "      <td>19.850841</td>\n",
       "      <td>0.0</td>\n",
       "      <td>83.00</td>\n",
       "      <td>97.0</td>\n",
       "      <td>110.00</td>\n",
       "      <td>146.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>226.0</td>\n",
       "      <td>103.938053</td>\n",
       "      <td>18.784051</td>\n",
       "      <td>49.0</td>\n",
       "      <td>93.00</td>\n",
       "      <td>105.0</td>\n",
       "      <td>117.00</td>\n",
       "      <td>147.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Language</td>\n",
       "      <td>0.0</td>\n",
       "      <td>735.0</td>\n",
       "      <td>86.576871</td>\n",
       "      <td>19.684342</td>\n",
       "      <td>48.0</td>\n",
       "      <td>72.00</td>\n",
       "      <td>86.0</td>\n",
       "      <td>100.00</td>\n",
       "      <td>139.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>237.0</td>\n",
       "      <td>98.717300</td>\n",
       "      <td>20.354492</td>\n",
       "      <td>48.0</td>\n",
       "      <td>83.00</td>\n",
       "      <td>100.0</td>\n",
       "      <td>114.00</td>\n",
       "      <td>163.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Receptive Language</td>\n",
       "      <td>0.0</td>\n",
       "      <td>716.0</td>\n",
       "      <td>89.607542</td>\n",
       "      <td>18.420979</td>\n",
       "      <td>50.0</td>\n",
       "      <td>75.00</td>\n",
       "      <td>90.0</td>\n",
       "      <td>103.00</td>\n",
       "      <td>138.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>229.0</td>\n",
       "      <td>99.593886</td>\n",
       "      <td>18.546673</td>\n",
       "      <td>50.0</td>\n",
       "      <td>87.00</td>\n",
       "      <td>101.0</td>\n",
       "      <td>114.00</td>\n",
       "      <td>149.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td rowspan=\"2\" valign=\"top\">Receptive Vocabulary</td>\n",
       "      <td>0.0</td>\n",
       "      <td>708.0</td>\n",
       "      <td>93.281073</td>\n",
       "      <td>17.992182</td>\n",
       "      <td>38.0</td>\n",
       "      <td>82.00</td>\n",
       "      <td>95.0</td>\n",
       "      <td>107.00</td>\n",
       "      <td>136.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1.0</td>\n",
       "      <td>230.0</td>\n",
       "      <td>101.591304</td>\n",
       "      <td>17.526472</td>\n",
       "      <td>42.0</td>\n",
       "      <td>91.00</td>\n",
       "      <td>103.0</td>\n",
       "      <td>113.00</td>\n",
       "      <td>159.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      count        mean        std   min  \\\n",
       "domain                mother_college                                       \n",
       "Articulation          0.0             485.0   84.853608  19.612918   0.0   \n",
       "                      1.0             130.0   87.723077  19.872468  40.0   \n",
       "Expressive Language   0.0             710.0   84.794366  18.938173  45.0   \n",
       "                      1.0             230.0   96.095652  20.233559  45.0   \n",
       "Expressive Vocabulary 0.0             702.0   96.188034  19.850841   0.0   \n",
       "                      1.0             226.0  103.938053  18.784051  49.0   \n",
       "Language              0.0             735.0   86.576871  19.684342  48.0   \n",
       "                      1.0             237.0   98.717300  20.354492  48.0   \n",
       "Receptive Language    0.0             716.0   89.607542  18.420979  50.0   \n",
       "                      1.0             229.0   99.593886  18.546673  50.0   \n",
       "Receptive Vocabulary  0.0             708.0   93.281073  17.992182  38.0   \n",
       "                      1.0             230.0  101.591304  17.526472  42.0   \n",
       "\n",
       "                                        25%    50%     75%    max  \n",
       "domain                mother_college                               \n",
       "Articulation          0.0             72.00   86.0  101.00  121.0  \n",
       "                      1.0             75.25   89.5  103.00  121.0  \n",
       "Expressive Language   0.0             69.00   85.0   99.75  141.0  \n",
       "                      1.0             81.00   98.0  111.00  140.0  \n",
       "Expressive Vocabulary 0.0             83.00   97.0  110.00  146.0  \n",
       "                      1.0             93.00  105.0  117.00  147.0  \n",
       "Language              0.0             72.00   86.0  100.00  139.0  \n",
       "                      1.0             83.00  100.0  114.00  163.0  \n",
       "Receptive Language    0.0             75.00   90.0  103.00  138.0  \n",
       "                      1.0             87.00  101.0  114.00  149.0  \n",
       "Receptive Vocabulary  0.0             82.00   95.0  107.00  136.0  \n",
       "                      1.0             91.00  103.0  113.00  159.0  "
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test_scores.groupby(['domain', 'mother_college']).score.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAw4AAAFOCAYAAAAvhUoIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3de1xUdf7H8fcICKKxZkEaqL+2m5culBbahVLLS4LKYKmQortdLEtX0zRgRROzVcvVUFurLUrblQRDzWxLXDM1K3dXU3HLDFdcuYmGgKjA/P7gx/ycQM+kcwNez8ejR8zhzJkP3xnPnPf5fs/3mCwWi0UAAAAAcAHN3F0AAAAAAM9HcAAAAABgiOAAAAAAwBDBAQAAAIAhggMAAAAAQw0+OFRWVio3N1eVlZXuLgUAAABotBp8cMjLy1OfPn2Ul5fn7lIAAACARqvBBwcAAAAAzkdwAAAAAGCI4AAAAADAEMEBAAAAgCGCAwAAAABDBAcAAAAAhggOAAAAAAwRHAAAAAAYIjgAAAAAMERwAAAAAGCI4AAAAADAEMEBAAAAgCGCAwAAAABDTg8OpaWlioiIUG5uriTpn//8px555BENHDhQkyZN0pkzZyRJ2dnZMpvN6tevnxISElRZWens0gAAAADYyanBYdeuXRoxYoRycnIk1YSIZ599Vi+++KI++ugjSdKqVaskSVOmTNH06dP1ySefyGKxKC0tzZmlAQAAAPgFnBoc0tLSlJSUpKCgIEnS1q1bFRoaqk6dOkmSEhMT9eCDD+rIkSOqqKhQaGioJMlsNmvDhg3OLA0AAADAL+DtzI3Pnj3b5vGhQ4fk7++viRMn6uDBg7r99ts1bdo07du3T4GBgdb1AgMDlZ+fX2d7JSUlKikpsVmWl5fnnOIBAAAAWDk1OPxcVVWVvvjiC61cuVJXX321EhIStGzZMt11110ymUzW9SwWi83jWqmpqUpJSXFlyQAAAADk4uBw5ZVX6tZbb1X79u0lSQMGDNDy5ctlNptVWFhoXa+oqMg6vOlccXFxioqKslmWl5en2NhY5xYOAAAANHEunY71nnvu0d69e3X06FFJ0qZNm9S1a1cFBwfL19dXO3fulCRlZmYqPDy8zvMDAgIUEhJi81/btm1d+ScAAAAATZJLexzatWunF198UWPHjtXp06fVuXNnTZ06VZI0f/58JSYmqrS0VF27dtWoUaNcWRoAAACACzBZLBaLu4u4FLm5uerTp482btyokJAQd5cDAAAANErcORoAAACAIYIDAAAAAEMEBwAAAACGCA4AAAAADBEcAAAAABgiOAAAAAAwRHAAAAAAYIjgAAAAAMAQwQEAAACAIYIDAAAAAEMEBwAAAACGvN1dAODJsrOztW/fPsP1ysvLJUn+/v4XXK9Lly7q3LmzQ2oDAABwJYID4ABlZWWSjIMDAABAQ0VwAC6gc+fOdvUQpKenS5Kio6OdXRIAAIBbcI0DAAAAAEP0ODQAjLMHAACAuxEcztHQD9AZZw8ADZs930Oe+h0EoPEjOFwEVx+gM84eAFCLk0QA3IXgcA4O0AEA7mTP9xDfQQDchYujAQAAABgiOAAAAAAwRHAAAAAAYIjgAAAAAMCQU4NDaWmpIiIilJuba7N8+fLlGjlypPXxf//7X8XGxqp///566qmnrDNGAAAAAPAMTptVadeuXUpMTFROTo7N8gMHDmjZsmXq2LGjddnMmTMVExOjgQMHavHixVqyZImmTJnirNIAnMOR88ZLzB3fGPCZAADUx2k9DmlpaUpKSlJQUJB12ZkzZzR9+nSNHz/euuzs2bP6+uuv1a9fP0mS2WzWhg0bnFUWgItQVlZGTyBs8JkAgKbHaT0Os2fPrrPslVdeUXR0tEJCQqzLjh8/rlatWsnbu6aUwMBA5efn17vNkpISlZSU2CzLy8tzYNVA0+Op88Zz1tt9PPUzAQBwL5fdAG7r1q06evSoXnjhBe3YscO63GKxyGQy2az788e1UlNTlZKS4tQ6ATQc3EEXAADXcVlwWLdunb7//nsNHjxY5eXlKioq0u9+9zvNmzdPJ0+eVFVVlby8vFRYWGgzvOlccXFxioqKslmWl5en2NhYV/wJAFyoKZz1plcFANCQuCw4zJkzx/rzjh07lJKSoj/+8Y+SpO7du2v9+vWKjIzUhx9+qPDw8Hq3ERAQoICAAJfUC9ey5wBKsv8gigMoNBb0qgAAPIXLgsOFJCUladq0aVq6dKnatWunV1991d0lwUNxEIXGpCn0qgAAGg+nB4esrKw6y8LCwhQWFmZ9HBwcrPfee8/ZpcCD2XMAJXEQBQAA4C4e0eMAAIArcX0JAPxyBAcAAOrB0EgAsEVwAAA0OVxfAgC/nNPuHA0AAACg8SA4AAAAADBEcAAAAABgiOAAAAAAwBDBAQAAAIAhggMAAAAAQwQHAAAAAIYIDgAAAAAMERwAAAAAGCI4AAAAADBEcAAAAABgiOAAAAAAwBDBAQAAAIAhggMAAAAAQwQHAAAAAIYIDgAAAAAMERwAAAAAGCI4AAAAADBEcAAAAABgiOAAAAAAwBDBAQAAAIAhpweH0tJSRUREKDc3V5K0cuVKRUREKDIyUi+88ILOnDkjScrOzpbZbFa/fv2UkJCgyspKZ5cGAAAAwE7eztz4rl27lJiYqJycHEnSjz/+qLfeeksZGRlq2bKlpk2bpvfff1+jR4/WlClTlJycrNDQUMXHxystLU0xMTHOLA8AALhYdna29u3bd8F1ysvLJUn+/v4XXK9Lly7q3Llzo64L8CRO7XFIS0tTUlKSgoKCJEnNmzdXUlKSWrVqJZPJpBtuuEH//e9/deTIEVVUVCg0NFSSZDabtWHDhjrbKykpUW5urs1/eXl5zvwTAACAi5WVlamsrMzdZdThqXUBruLUHofZs2fbPA4ODlZwcLAkqbi4WCtWrNCcOXNUUFCgwMBA63qBgYHKz8+vs73U1FSlpKQ4s2QAAOBEnTt3Njwbn56eLkmKjo52RUmSPLcuwJM4NTicT35+vh577DFFR0crLCxMO3fulMlksv7eYrHYPK4VFxenqKgom2V5eXmKjY11es0AAABAU+by4PDDDz/oscce08iRI/Wb3/xGktS2bVsVFhZa1ykqKrIObzpXQECAAgICXFYrAAAAgBounY61tLRUv/3tbzVhwgRraJBqhjD5+vpq586dkqTMzEyFh4e7sjQAAAAAF+DSHodVq1apqKhIb7/9tt5++21JUu/evTVhwgTNnz9fiYmJKi0tVdeuXTVq1ChXluY2n3/+uU1vy6Wo3U7tGMxLERgYSHgDAACAlUuCQ1ZWliRp9OjRGj16dL3rdOrUSatWrXJFOR6lsLBQhYWFNheHX6yWLVs6oCI5LMgAAACg8XDLxdGwFRgY6FEzNDiixwIAAACNi0uvcQAAAADQMBEcAAAAABhiqBKaLC5MR1PkqM+9Iz/zEp97AGgICA5osrgwHU2Roz73jvrMS03nc09oA9DQERzQpHFhOpoiPvfuQWgD0NARHAAAcBFCG4CGjIujAQAAABiixwGAyzHWGwCAhqfJBAdPnUHnxIkTat269SVvB2hIGOsNAEDD02SCQ+2BSps2bS55Wy1atJAkVVVVXdJ2iouLL7kWoKFirDcAAA1LkwkOktSmTRsNGDDA3WVYffzxx4QHAAAANAhcHA0AAADAEMEBAAAAgKEmNVQJAND4eeqsXUyGAaChIzgAABoVR02G4aiJMCQmwwDQOBAcAACNDpNhAIDjERzcrKqqSoWFhR41FWRhYaFD58cHAABAw8fF0QAAAAAM0ePgZl5eXtwICwAAAB6PHgcAAAAAhuhxgNM5ampEybHTIzI1IgBX4po29/DU7yBJCgwMVHh4uEO2BbgCwQFO56ipESXHTY/I7CYA0DTUfgcFBgZe8rYcGbIcFWYAV3J6cCgtLdXw4cP1+uuvKyQkRNu2bdOcOXN0+vRpDRgwQBMnTpQkZWdnKyEhQWVlZerevbtmzpwpb29yTWPB1IgAmjquaXMfT2t3qem0PRoXpx6Z79q1S4mJicrJyZEkVVRUKD4+Xu+9957atWunJ598Ups3b9Z9992nKVOmKDk5WaGhoYqPj1daWppiYmKcWR6aOIYNAIDjeOqQIEf1NgBwcnBIS0tTUlKSnn/+eUnS7t271bFjR7Vv316SFBkZqQ0bNui6665TRUWFQkNDJUlms1mLFi2qExxKSkpUUlJisywvL8+ZfwIAXBRHHUQ5eky1o7fnCATmxsETh6VK0tmzZy95GwBqODU4zJ492+ZxQUGBTeoPCgpSfn5+neWBgYHKz8+vs73U1FSlpKQ4r2A0KU1h2ICnHrw2hQvTHXUQ5cgDqOLiYlVVVcnLy+uStwXUx9OGpUrS8uXL3V0C0Gi49CKC6upqmUwm62OLxSKTyXTe5T8XFxenqKgom2V5eXmKjY11XtFAA+apB69NhacdRNVe29PYAzMAwDlcGhzatm1rc/azsLBQQUFBdZYXFRUpKCiozvMDAgIUEBDgklqBxsJTD14BAEDD4tLgcOutt+rHH3/UoUOHFBISonXr1ik6OlrBwcHy9fXVzp071a1bN2VmZjKvMdCIcWE6AAANj0uDg6+vr15++WU9++yzOn36tO677z71799fkjR//nwlJiaqtLRUXbt21ahRoxz62uXl5SotLdXHH3/s0O1eimPHjqm6utrdZQAAAACGXBIcsrKyrD/37NlTa9asqbNOp06dtGrVKleUA8DNmsKF6QAANDZN5g5r/v7+8vX1Zaw3AAAAcBGaTHAAADQNDE0FAOdo5u4CAAAAAHg+ehwAAI0KQ1MBwDnocQAAAABgiB4HD+Co+ezLy8sl1Zxtu9R6AgMDL7keAAAANB4EBzfz8fFR69atHbKtsrIySZceHAIDAwkOAAAAsEFwcLPWrVs7bC772l4LT5obHwAAAI0DwQEAABdxxNBURw1Lra2HHmYA9iI4wOmYUx0AHDc01VHDUiWGpgL4ZQgOAAC4gKOGpjIs9Zeprq522CQkjlRYWKiWLVu6uwzgFyE4wOmYUx0AAKDhsys4lJWVaf78+Tp48KAWLlyoV199VVOnTiUpAwAAj9asWTMFBgZ6XA+Np/WAAPaw6wZwycnJCggI0LFjx+Tr66vS0lJNnz7d2bUBAAAA8BB29ThkZ2drzpw52rx5s1q0aKH58+crIiLC2bUBTsfN9wAAAOxjV3Bo1sy2Y6KqqqrOMqCh4eZ7AAAA9rMrONxxxx2aN2+eKioqtGXLFq1YsUJhYWHOrg1wKm6+B2diGmI0NZ74mZekyspKa68wgEtjV7fB5MmT5e/vr8suu0wLFizQjTfeqOeff97ZtQEAAADwEHb1OCxatEjPPfecxo0b5+x6ADiQJ54BbCpnvZmGGE2NJ37mJWn58uUOuVkeADt7HP7+9787uQwAAAAAnsyuHoeQkBD95je/0e23325z74YxY8Y4rTAAl84TzwBy1hsA0FA9/fTTWrJkiTIyMvT9999r6tSpLq+h9rVjY2M1fvx4ZWRkuOy17QoOtTPPHDlyxKnFOFtxcbFDhmycOnVKktSiRYtLrocZdAC4miOmIXbUFMS19bAvBNAQ7Ny5090luJVdwWHOnDmSaoJDZWWlOnbs6NSinMGRX0q1waFVq1aXtB2m3gTgao6ahthRUxBLztkXOuJEkaNOEtXWw/4e8BwZGRnasmWLjh49qoKCAr3wwgtauXKlDhw4oMTERIWEhGj69Ok6deqUbrzxRs2aNUuLFy/WyZMnNX78eN1///3Kzs7W8OHDVVBQoGnTpqlv377Kz89XYmKiioqKdMUVV+ill17SmTNn9PTTT8vb21thYWHn7aX44IMP9Pbbb8tkMmnMmDEaOnSotm/frj/84Q+qqqpSWFiY4uPj633ugQMHNHPmTJWWlqpDhw6aPXu2WrVqpddee03r1q1TSEiIjh07ppSUFLVu3VoJCQk6fPiwWrRooRdffFHXXnutXe1mV3A4dOiQnn76aRUUFKi6ulqXX365/vSnP9n9Ij+XmZmpZcuWSZLCw8M1depUZWdnKyEhQWVlZerevbtmzpwpb2+7yrNLeHi4w7bF1JsAGipHTUPsyftBRx2gO+okkcSJIsATHThwQOnp6Vq/fr1mz56t9evX69tvv9XSpUt14sQJzZgxQ6GhoZo7d67+/Oc/a/Lkyfrggw+0aNEiZWRk6NixY0pPT9fhw4c1btw49e3bV7Nnz9aECRN000036dNPP9Urr7yiZ599Vjk5Ofr888/Pe+ImLy9Pb775ptLT02UymTRy5Ej169dPiYmJeueddxQSEqKJEydqzZo19T5/5syZmjt3rtq1a6d33nlHf/7zn9WrVy9t2bJF69at048//qghQ4ZIkhYvXqyBAweqb9++2rt3r5KSkrR8+XK72syuI/MXX3xRjz32mKKioiTVfGHMnDlT7777rl0vcq5Tp05p9uzZ2rBhgwICAjRixAht27ZNL730kpKTkxUaGqr4+HilpaUpJibmF28fQMPAcBk4i6NOFHlyOAJw6cLCwtS8eXO1a9dOXbp0kb+/v4KDg1VSUqJTp04pNDRUkjRo0CC9/PLLdZ7fo0cPNW/eXNdee61OnDghSfr666916NAhSVJ1dbUCAgIk1VwvfKHe3n/9618KCwuznqjIyMjQv//9b3Xs2FHt27eXJEVGRiorK0vdunWzeW5paal27dqlsWPHSqq5d0mnTp309ddf64EHHpCPj49uuOEG3XjjjZKkHTt26IsvvtDixYutz7eXXcHh2LFj1tAg1exE33nnHbtf5FxVVVWqrq7WqVOn5O/vr8rKSnl7e6uiosL6BpnNZi1atIjgADRSTWW4DADAc/n4+Fh/PneUS31ThldVVdVZdu5zLBaL9f8ZGRny8vLS2bNndfLkSZWXl8vPz++Ctfx8lM3hw4dVUVFhs8xisdRbR3V1tQIDA5WZmSmp5iT9mTNntHr1amtdkmQymax/y5/+9CddffXVkqT8/PwL1mZTpz0rVVVV6cSJE9Yv+kuZEaVVq1aaMGGCBgwYoBYtWuiOO+6Qj4+PzRd2YGBgvX9ESUmJSkpKbJbl5eVddC0A3KMpDJcBADRMzZo1U/PmzfWvf/1LoaGhyszM1G233WbXc2+55RatXr1aQ4cO1V//+lft3r1bEyZMMHzeTTfdpFdeeUVlZWUymUx67LHH9N577yknJ0eHDx9WSEiI1q1bp549e9Z5bkBAgFq0aKHt27erZ8+eWrhwofz8/PTAAw8oOTlZv/3tb3XkyBF99913MplMuv322/XBBx9owoQJysrK0ptvvqn333/frr/PruDw6KOPatiwYRowYIBMJpPWr1+vuLg4u17g5/bv36/09HRt2rRJl112mSZPnqytW7daU5BUk6jOfVwrNTVVKSkpF/W6AAAAgD1efvllzZgxQ2VlZbruuuv00ksvSZJuu+02jR49WoMGDar3eb///e+VkJCg1NRUtW7dWq+88orOnDlj+Hpt27bV448/ruHDh8tisWjcuHEKCgrSrFmz9Mwzz+j06dO68847FR0dXe91DnPnzlVSUpIqKioUHBysV155RS1bttQ999yjIUOGKCQkRFdeeaV8fX01fvx4xcfHKzIyUr6+vvUOwzofu4LDsGHD1LFjR23ZskXV1dWaMWNGvYnHHl988YV69uypK664QlLNsKS33npLhYWF1nWKiooUFBRU57lxcXE2Q6akmh6H2NjYi6oFAAAATY/ZbLb+HBYWprCwMEk11yLU3hfhgw8+qPO8119/vd7t7dixQ5LUvn37eq8BtudeC2az2aYuSbr77rutQ5Dqq712u126dKlT78GDBxUcHKx169YpPz9fw4YNU5s2bdSsWTMtXbrUsJ762BUc8vPztWHDBs2YMUMHDx7U/Pnzdd11113UeOBOnTpp3rx5Ki8vV4sWLZSVlaU777xTn3zyiXbu3Klu3bopMzOz3ovbAgICrBeZAAAAAA3Ju+++W+/EIE899ZT69+/v0Ndq166d5s2bp2XLlsnb21sJCQlq1qzZJW3TruAwdepU9e7dW5IUHBysO++8U/Hx8XrjjTd+8Qvec8892rdvn8xms3x8fHTzzTfriSee0IMPPqjExESVlpaqa9euGjVq1C/eNgAAAOCpRo0a5bJj3BYtWlx0z8L52BUcjh8/bv0jfX19NXr0aH344YcX/aJPPPGEnnjiCZtlnTp10qpVqy56mwAAAACcx67+iqqqKptZjoqKimymdwIAAADQuNnV4zB69GgNGTJE9957ryRp+/btev75551aGAAAgCM44oaTEjedBOwKDkOHDtVNN92kL7/8Ul5eXurQoYPuv/9+J5eGxqS4uFgff/zxJW/n1KlTkmrG7V1qPeywAaDx8/HxUWVlpUO2xU0n0dTZFRymT58uqeaCjtGjR+vee+9VfHy8XnvtNacWh8bBkTvG2uBQe0v2i8UOGwCahoCAAHl5eXHTSVy0tWvX1rmLsyP4+fkpMjLS7hqWLl2qyspKxcXF1bkVQXZ2thISElRWVqbu3btr5syZde5G7Qh2bXHPnj1atWqVli1bpqioKD333HN15pkFzqe+qXUvFjttAADgShUVFQ6fKlWSNmzYYNd6+fn5WrBggTIyMtS8eXMNHz5cYWFhuu6666zrTJkyRcnJyQoNDVV8fLzS0tIUExPj8JrtujjaYrGoWbNm2rp1q3r06CFJTkleAAAAAP7ftm3b1KNHD7Vu3Vr+/v7q16+fTeg4cuSIKioqFBoaKqnmBnH2hpJfyq4ehw4dOujxxx9Xbm6u7rzzTj333HPq1KmTUwoCgMbCEdf2OOq6ntp6GKIHZ/K069kkPvdo+AoKCmw+w0FBQdq9e/d5fx8YGGgzG6oj2RUc5syZo08//VTdunWTj4+PunfvriFDhjilIABoDBx1oOKo63okru2Bc3ni9WwSn3s0fNXV1TKZTNbHFovF5rHR7x3JruDg7++vwYMHWx+PGDHCKcUAQGPhqGt7uK4HDQXXswHO0bZtW33zzTfWx4WFhQoKCrL5fWFhofVxUVGRze8dya5rHAAAAAC43l133aXt27eruLhYp06d0t/+9jeboB4cHCxfX1/t3LlTkpSZmenQIH8uggMAAADgoa666ipNnDhRo0aN0pAhQxQREaFbbrlFjz/+uL799ltJ0vz58zVnzhz1799f5eXlGjVqlFNqcfwErwA8ChfoAgBw8fz8/JwyS5Gfn5/d60ZGRta558Mbb7xh/blTp05atWqVw2o7H4ID0IhxgS4AAJfG3pu0NQUEB6AR4wJdAADgKFzjAAAAAMAQwQEAAACAIYIDAAAAAEMEBwAAAACGCA4AAAAADDGrEgAAAHAeH330kSoqKhy+XT8/Pw0cONCudUtLSzV8+HC9/vrrCgkJsflddna2EhISVFZWpu7du2vmzJny9nbOIT7BAQAAADiPiooKp0xHXjvVuZFdu3YpMTFROTk59f5+ypQpSk5OVmhoqOLj45WWlqaYmBgHVvr/GKoEAAAAeKi0tDQlJSUpKCiozu+OHDmiiooKhYaGSpLMZrNT7nJdix4HAAAAwEPNnj37vL8rKChQYGCg9XFgYKDy8/OdVotbehyysrJkNps1YMAAJScnS5K2bdumyMhI9e3bVwsWLHBHWQAAAECDUV1dLZPJZH1ssVhsHjuay4PD4cOHlZSUpCVLlmjNmjXat2+fNm/erPj4eC1ZskTr16/Xnj17tHnzZleXBgAAADQYbdu2VWFhofVxUVFRvUOaHMXlweHTTz/VQw89pLZt28rHx0cLFixQixYt1LFjR7Vv317e3t6KjIx06vgsAAAAoKELDg6Wr6+vdu7cKUnKzMxUeHi4017P5dc4HDp0SD4+Pho7dqyOHj2q+++/X9dff73N+KygoKB6x2eVlJSopKTEZlleXp7Ta3a37Oxs7du3z3C92sRpdJV+ly5d1LlzZ4fUBgAA0Jj5+fnZPQPSL93uxXr88cc1fvx43XzzzZo/f74SExNVWlqqrl27atSoUQ6s0pbLg0NVVZW++eYbvffee/L399dTTz0lPz8/u8ZnpaamKiUlxZXlNigtW7Z0dwkAAACNir33WnC2rKws689vvPGG9edOnTpp1apVLqnB5cHhyiuvVM+ePdWmTRtJ0gMPPKANGzbIy8vLuk5hYWG947Pi4uIUFRVlsywvL0+xsbHOLdrNOnfuTA8BAAAA3MrlwaFXr16aOnWqSkpK1LJlS23ZskX9+/fXsmXLdOjQIYWEhGjdunX13mgjICBAAQEBri4ZAACXsGdoKsNSAbiLy4PDrbfeqscee0wxMTE6e/as7r77bo0YMUK//vWv9eyzz+r06dO677771L9/f1eXBgCAx2NYKgB3ccsN4IYOHaqhQ4faLOvZs6fWrFnjjnIAAPAIDE0F4MnccgM4AAAAAA0LwQEAAACAIYIDAAAAAEMEBwAAAACGCA4AAAAADBEcAAAAABgiOAAAAAAwRHAAAAAAYIjgAAAAAMAQwQEAAACAIW93FwBIUnZ2tvbt22e4XmFhoSQpPT39gut16dJFnTt3brR1AQAAuBrBAQ1Ky5Yt3V1CvTy1LgAAAEchOMAjdO7c2SPPxHtqXQAAAK5GcAAAN7FnKJy9w+Ak1w7Rc0ddjtQU/kYAcDSCAwB4ME8dBuepdTlSU/gbAeCXIDgAgJt46lA4T63LkZrC3wgAjsZ0rAAAAAAMERwAAAAAGGKoEtDEeepFop5aFwAATRXBAYAhT71I1FPrAgCgMSI4AE2cp14k6ql1AQDQVBEczmHP0AjJ/uERDI0AAABAY0FwuAgMjwAAAEBTQ3A4B0MjAAAAgPq5dTrWP/zhD5o2bZqkmmFCZrNZ/fr1U0JCgiorK91ZGgAAAIBzuC04bN++XatXr7Y+njJliqZPn65PPvlEFotFaWlp7ioNAAAAwM+4ZajSiRMntGDBAo0dO1b79+/XkSNHVFFRodDQUEmS2WzWokWLFBMTY/O8kpISlZSU2CzLy8tzWd0AAODSOPIeLY6chMRT6wI8iVuCw/Tp08Me7yMAABjMSURBVDVx4kQdPXpUklRQUKDAwEDr7wMDA5Wfn1/neampqUpJSXFZnQAAwPU8dRIST60LcBWXB4cPPvhA7dq1U8+ePZWRkSFJqq6ulslksq5jsVhsHteKi4tTVFSUzbK8vDzFxsY6t2gAAOAQnjoRiafWBXgSlweH9evXq7CwUIMHD9ZPP/2k8vJymUwma/efJBUVFSkoKKjOcwMCAhQQEODKcgEAAADIDcHh7bfftv6ckZGhr776SnPmzFFERIR27typbt26KTMzU+Hh4a4uDQAAAMB5eMx9HObPn6/ExESVlpaqa9euGjVqlLtLAgAAAPB/TBaLxeLuIi5Fbm6u+vTpo40bNyokJMTd5QAAAACNkltvAAcAAACgYSA4AAAAADBEcAAAAABgiOAAAAAAwBDBAQAAAIAhggMAAAAAQwQHAAAAAIYIDgAAAAAMERwAAAAAGCI4AAAAADBEcAAAAABgiOAAAAAAwBDBAQAAAIAhggMAAAAAQwQHAAAAAIYIDgAAAAAMERwAAAAAGCI4AAAAADBEcAAAAABgiOAAAAAAwBDBAQAAAIAhggMAAAAAQwQHAAAAAIbcEhxSUlI0cOBADRw4UHPnzpUkbdu2TZGRkerbt68WLFjgjrIAAAAAnIfLg8O2bdv0xRdfaPXq1frwww+1d+9erVu3TvHx8VqyZInWr1+vPXv2aPPmza4uDQAAAMB5eLv6BQMDAzVt2jQ1b95cknTttdcqJydHHTt2VPv27SVJkZGR2rBhg+677z6b55aUlKikpMRmWV5enmsKBwAAAJowlweH66+/3vpzTk6OPv74Yz366KMKDAy0Lg8KClJ+fn6d56ampiolJcUldQIAAAD4fy4PDrW+//57Pfnkk3r++efl5eWlnJwc6+8sFotMJlOd58TFxSkqKspmWV5enmJjY51dLgAAANCkuSU47Ny5U+PHj1d8fLwGDhyor776SoWFhdbfFxYWKigoqM7zAgICFBAQ4MpSAQAAAMgNF0cfPXpU48aN0/z58zVw4EBJ0q233qoff/xRhw4dUlVVldatW6fw8HBXlwYAAADgPFze4/DWW2/p9OnTevnll63Lhg8frpdfflnPPvusTp8+rfvuu0/9+/d3dWkAAAAAzsNksVgs7i7iUuTm5qpPnz7auHGjQkJC3F0OAAAA0Chx52gAAAAAhggOAAAAAAwRHAAAAAAYIjgAAAAAMERwAAAAAGCI4AAAAADAEMEBAAAAgCGCAwAAAABDBAcAAAAAhggOAAAAAAwRHAAAAAAYIjgAAAAAMERwAAAAAGCI4AAAAADAEMEBAAAAgCGCAwAAAABDBAcAAAAAhggOAAAAAAwRHAAAAAAYIjgAAAAAMERwAAAAAGCI4AAAAADAEMEBAAAAgCGPCg5r167VQw89pL59+2rFihXuLgcAAADA//F2dwG18vPztWDBAmVkZKh58+YaPny4wsLCdN1117m7NAAAAKDJ85geh23btqlHjx5q3bq1/P391a9fP23YsMHdZQEAAACQB/U4FBQUKDAw0Po4KChIu3fvtlmnpKREJSUlNsvy8vJcUh8AAADQlHlMcKiurpbJZLI+tlgsNo8lKTU1VSkpKa4uDQAAAGjyPCY4tG3bVt988431cWFhoYKCgmzWiYuLU1RUlM2yvLw8xcbGuqRGAAAAoKnymOBw11136bXXXlNxcbFatGihv/3tb5o1a5bNOgEBAQoICHBThQAAAEDT5THB4aqrrtLEiRM1atQonT17VkOHDtUtt9zi7rIAAAAAyIOCgyRFRkYqMjLS3WUAAAAA+BmPmY4VAAAAgOciOAAAAAAwRHAAAAAAYIjgAAAAAMAQwQEAAACAIYIDAAAAAEMEBwAAAACGPOo+DhejqqpKkpSXl+fmSgAAAJyjbdu28vZu8IdtaOAa/CewsLBQkhQbG+vmSgAAAJxj48aNCgkJcXcZaOJMFovF4u4iLkVFRYX27NmjwMBAeXl5ubucXywvL0+xsbFasWKF2rZt6+5ymhTa3j1od/eh7d2HtnefxtL29DjAEzT4T6Cfn5+6d+/u7jIuWdu2bTmT4Ca0vXvQ7u5D27sPbe8+tD1w6bg4GgAAAIAhggMAAAAAQwQHAAAAAIa8ZsyYMcPdRTR1vr6+CgsLk6+vr7tLaXJoe/eg3d2Htncf2t59aHvAMRr8rEoAAAAAnI+hSgAAAAAMERwAAAAAGCI4XEBubq5uuukmDR482Oa/FStWuLs0JSQk6Ntvv3V3GU738/cgMjJSvXv31qJFi5z+2ocPH1Z8fLwk6dtvv1VCQoLTX9Pd+Mw3PDNnztTgwYP10EMP2bx36enp9a5/6NAhJSYmXnCbhw4d0oMPPuiMcj0a+xv3+e6773TjjTfqk08+Oe86I0eOtP48ePDgi3qd3r17Kzc397y/P3nypMaNGydJys/P1+OPP35RrwM0Vg3+BnDOFhQUpMzMTHeXUcfs2bPdXYLL/Pw9yM/PV79+/TRw4EBde+21Tnvd//73vzp8+LAk6eabb9bNN9/stNfyJHzmG5akpCRJNQe9o0aNMnzvjhw5csEDp6aO/Y17pKenq3///lq5cqX69etX7zpfffWV9Wdn7aN++uknZWdnS5KuuuoqvfHGG055HaChosfhIuzdu1d33323iouLdeLECfXq1UvZ2dl67bXX9MILL+iRRx7Rgw8+qDfffFOSlJGRoZEjRyoyMlKvvvqqioqK9PTTT8tsNis6Olrbtm2TJG3fvl1ms1lms1ljxoxRcXGxSktL9cQTT1iXb9y4UVLNmZcdO3bomWeesTlDYzabtW/fPh06dEhjxoxRVFSURowYoX379rm+oZyksLBQFotFLVu21LJlyxQVFaVBgwZp7ty5qr3W/5133lG/fv300EMPad68eZJ03nY/3/uWnJysPXv2aObMmdqxY4dGjhyp/fv3KzIy0lpLVlaWnnrqKUk6by2NAZ/5hqesrEyTJk1SRESEIiMjtWbNGkk1n+tdu3YpOTlZZ8+eVXx8vB555BH17t1bzzzzjE6fPu3myj0L+xvnO3v2rNauXavf/e532rt3r/7zn/9Iqukd+N3vfqd+/fqpdgLIhx9+WJJ04403SpJOnDihcePGacCAARo8eLC2b99u83upZn80bdo0m9csLS3V+PHjNWzYMPXq1Uvx8fGyWCxKTk5WQUGBxo0bp9zcXPXu3VtSzfv55JNPKjIyUlFRUfr8888l1byfiYmJGjlypHr37q2lS5c6r6EAD0CPg4GCgoI6XaJz587VsGHDNHfuXJ09e1YjRoxQ586d9dlnn2nPnj3661//qurqapnNZvXs2VNSzVmr9evXy9vbWxMnTlR0dLT69OmjgoICxcTE6MMPP9SSJUs0Y8YM3XLLLXrjjTe0b98+FRYWKjg4WMuWLVN2drbWrFmjPn36WGsZPHiw1q5dq379+iknJ0enT59Wly5dNHz4cE2fPl1dunTRgQMHNG7cuAt2AXuy2vfg9OnTOn78uG6++WalpKTou+++0549e7Rq1SqZTCZNmTJFa9as0TXXXKP3339f6enpatGihR577DHt2bNHb731Vr3tLqne9y0xMVEpKSlKSkrSjh07JEmdOnWSyWTSd999pxtuuEEfffSRBg0apM8//7zeWi62O92d+Mw3DosWLVJgYKBeffVVHTt2TA8//LC6dOmixMRELVu2TImJifryyy/l7++vtLQ0VVdXKzY2Vlu2bNH111/v7vLdhv2N623evFlXX321rrnmGj3wwANauXKlpkyZIkkKDw/XH//4R0nSX/7yF33wwQc2z124cKE6dOigxYsX69///remT59u3QddyN///nd17txZixYt0pkzZzRw4EDt3btXiYmJGjVqlBYvXmzTMzdr1iz16NFDY8aM0eHDhzVixAjr+/nvf/9bK1as0MmTJ/XAAw8oNjZWAQEBjmoewKMQHAycb9jGr3/9a0VHR8vPz896hkmSIiIi1LJlS0k1Z0u+/PJLXX755erSpYu8vWuae9u2bTp48KB13GxlZaUOHz6sPn366JlnntEDDzygPn366O6771ZOTo5effVV5efn6/7777eOvax133336cUXX1RpaanWrVunQYMGqaysTHv27NELL7xgXa+8vFzHjx/X5Zdf7vA2crba96C6ulovv/yyfvjhB919992aN2+edu/eLbPZLEmqqKjQ1VdfraKiIvXq1UuXXXaZpJqzgdL5212q/3276aab6q1n0KBB+uijj9ShQwd9/fXXeumll/THP/6x3loaIj7zjcOXX36p+fPnS5KuuOIK9erVSzt27NA111xjXadHjx5q06aNVqxYoYMHD+rw4cMqLy93V8kegf2N66WnpysiIkKS9NBDD2ny5MmaMGGCJOnWW2+94HO//vpr6+f8xhtv1MqVK+16zYiICO3evVvvvPOODh48qBMnTqi8vFytW7eud/0vv/xSycnJkqT27dvr1ltv1a5duyRJYWFhat68ua644gq1bt1aJ0+eJDig0SI4XKSTJ0+qrKxMZWVlOnHihNq0aSNJ8vLysq5TXV1tfezn52ezPDU11bqDKigo0BVXXKHOnTurV69e2rRpk/VL6qmnntLHH3+sLVu2aNOmTfrzn/+s9evXW7fVvHlz9erVS1lZWdqwYYP+9Kc/qbq6Ws2bN7c5+MvLyzvvDrGhaNasmZ5//nkNGTJEb731lqqqqhQXF6cxY8ZIkkpKSuTl5WU9C1crPz9fLVq0OG+7f/bZZ+d93+oTGRmpuLg4derUSffcc498fX3PW0tjwme+YamurrZ5bLFYVFVVZbPs008/1eLFizVq1CiZzWbrsBywv3GVY8eOacuWLdq7d6/effddWSwWlZSU6NNPP5Ukwxu2eXt727T/Dz/8YA3HFotFJpNJlZWVdZ733nvv6ZNPPtEjjzyiu+66S999990FP/s//925/57OrdFkMvFvCI0a1zhcpJkzZ+rRRx9VTEyMZs6caV3+2Wef6cyZM/rpp5+0adMm3XPPPXWe26NHD73//vuSpAMHDigyMlKnTp3Sww8/rLKyMo0ePVqjR4/Wvn37tHz5cr322msaMGCAkpKSrGPAzzV48GC9/fbbat26tYKDg3XZZZfpf/7nf6wHUVu3blVsbKwTW8N1vL299fzzz2vJkiXq0qWLMjMzVVZWpsrKSuvQlO7du2vz5s3W5c8995z27Nlz3naX6n/fvLy86v3Cueqqq9SuXTstW7ZMgwYNklTzntZXS2PCZ75h6dGjh1atWiVJKi4uVlZWlu688055eXlZD3i2bt2qgQMHymw2q1WrVvr666/rBI6mjP2N82VmZqpHjx76/PPPlZWVpU2bNmns2LH661//Wmfd+tqoe/fu+uijjyTVhIbHH39cJpNJl19+ub7//ntZLBZlZWXV2dbWrVs1bNgwDRo0SKdPn9b+/ftVXV0tb2/vet+Hc/89HT58WP/4xz8UGhrqiCYAGhR6HAzUN9777Nmz8vPz06uvviqLxaLo6GjrGVFfX1/FxMSotLRUTz75pK677jrt3r3b5vmJiYmaPn269aK3uXPnqlWrVpo0aZKmTZsmb29v+fv7Kzk5WVdeeaUmTZqkyMhIeXl5acqUKXW6QLt166aTJ09qxIgR1mXz5s3TjBkz9Oabb8rHx0cLFiywOSvTkIWHh+u2227TN998o759++qRRx5RVVWV7r33XkVFRclkMunRRx/V8OHDVV1drQcffFB33XWXrr322nrbXar/fTt+/LhOnjypKVOmaOjQoTY1DB48WAsWLNCdd94pqWa4wf79++vU0hDxmW8cxo8frxkzZigyMlJVVVUaN26cOnXqpOLiYhUXF2vatGmKi4uzjo/38fFRt27dlJubq9tvv93d5XsM9jfOtXr1ak2cONFmWWxsrN58801re9Xq06ePBg8erIyMDOuy8ePHKzExUYMGDZK3t7fmzp0rk8mk5557TmPHjtWVV16pbt266fjx4zbbiouL04wZM7Rs2TK1atVKt912m3Jzc9W9e3ddffXVGjlypObMmWNdPyEhQdOnT7e+dnJysoKCghzdHIDHM1noU3OY1157TZL07LPPurkS/BK8bxePtgN+Gf7NAGjIGKoEAAAAwBA9DgAAAAAM0eNQjxEjRlgvtqpVXl6usLAwFRcXX/L2z72pjL1qb37V1NTeCAmuR9u7D23vOuzvAcB+BId6REdHa+3atTbL/va3vyksLMw6BSUAoOFjfw8A9iM41GPAgAH6xz/+oRMnTliXrVmzRtHR0frXv/6lhx9+WIMGDVJcXJwOHTokScrOztbDDz+syMhIPfroo8rLy1NlZaUSExM1bNgw9enTR08//bQqKiokSadPn9aECRM0aNAgPfPMM/rpp58k1cyWUXu3yvrOOp5vm7m5uerfv79GjBihMWPGKCYmRlu3bpVUM9903759lZ+f7/S2c4ULtcGQIUM0ZcoURUREKC4uzvoerl+/Xv3795fZbFZCQoKmTZsm6fzt/dVXX2nEiBGKiopSnz599Nlnn0mquTfAo48+qsjISD333HMKDw+XJJWVlWnq1Kkym80aPHiw1q1b5+pmcQna3n1oe+dgfw8A9iM41KNly5bq06ePNmzYIKnmhj4//vijevbsqUmTJun3v/+91qxZo+HDh2vSpEmSpMmTJ+vpp5/W2rVr9dBDDyk1NVX//Oc/5ePjo5UrV+rTTz/VyZMntXnzZkk1N70ZOXKk1qxZo/bt22vx4sV21Xahbf7444+aN2+e3n77bUVHR1vntP/mm2/UoUMHXXXVVY5uKre4UBvs379fY8aM0bp16xQQEKC1a9equLhYL730klJTU7Vq1Srrl/aFLF++XMnJyVq9erWSk5O1cOFCSdLs2bM1YMAArV27Vv3797d+OS9dulRdu3ZVRkaGVqxYoddff916l9jGhLZ3H9reOdjfA4D9uI/DeZjNZi1cuFDDhw/X2rVrNWjQIOXk5CggIEC33HKLpJozVdOnT9eRI0dUWFioXr16SZJiYmKs22ndurVWrFihgwcPKicnR+Xl5ZKka665Rt27d5dUM0d37ZlAI3fcccd5t3nFFVcoJCTEWtuCBQtUXl6u1atXy2w2O6ZhPIBRG3Tp0kWSdP311+unn37SN998o9tuu836RTpkyBDrmdTzmTdvnjZt2qQNGzZo165dKisrk1Rz06Daub0ffPBB6/0Ftm3bpoqKCqWnp0uqGSP9/fffq3379o5vADei7d2Htnce9vcAYB96HM7jjjvuUGFhoY4ePWrttq7vjqq1k1Kde6Op06dP6/Dhw9q4caMmT54sPz8/mc1m3XHHHdb1vb29bbbx88eS6r175YW26efnZ13P399f4eHh+uSTT/Tll1+qT58+l9IcHuVCbeDr62tdz2QyyWKxqFmzZhe8G2597R0TE6Pdu3frpptu0tixY63Lvby8VN9EZNXV1Zo3b54yMzOVmZmptLQ03XvvvZf8t3oa2t59aHvnYX8PAPYhOFzAkCFDtHTpUv3qV79Shw4d9Otf/1onTpyw3hV3/fr1uvrqqxUcHKyrrrpKX3zxhSQpMzNTCxcu1Pbt2zVgwABFR0crICBAO3bsUFVVlSTphx9+0L59+yRJ6enpuuuuuyRJl19+uQ4cOCCp5kvj5y60zZ+Ljo7WggULdO+999ocWDR0v6QNJOn222/Xt99+q4KCAlksFq1fv976xV9fe584cUI5OTmaMGGCwsPDtXHjRuv2e/bsab2QcvPmzSopKZEk9ejRQ3/5y18k1dx5edCgQTp69KhzGsCNaHv3oe2di/09ABhjqNIFmM1m9e7dW7Nnz5YkNW/eXAsWLNCsWbN06tQp/epXv9KCBQsk1XTxz5gxQ/PmzdPll1+uuXPn6vjx45o8ebI++ugj+fj46Pbbb7deCNehQwctXrxY//nPf3TDDTdo4sSJkqTx48dr1qxZSklJ0T333FOnpocffvi82/y5bt26yWQyKTo62hnN4zK1Qy5q3XLLLdqxY4ddbSBJbdq0UWJion7zm9+oefPmCgkJsQ61qK+9W7duraFDh2rgwIHy9vZWjx49VFFRofLyciUkJGjq1KlKS0tTp06drNt55plnNGPGDEVERKiqqkpTpkxRhw4dnNgqrkHbuw9t71rs7wHADhY0StXV1Zb9+/dbBg8e7O5S3K64uNiycOFCS1VVlcVisVhmzZpleffddy9qW6mpqZbvv//eYrFYLHv27LFERUU5rM7GiLZ3H9q+6WB/D8BV6HFopFJTU/Xmm29aZ0Vpylq3bq2SkhJFRETIy8tLXbt21SOPPHJR2+rYsaMmTZqkZs2aydfXV7NmzXJwtY0Lbe8+tH3Twf4egKuYLJZ6rngDAAAAgHNwcTQAAAAAQwQHAAAAAIYIDgAAAAAMERwAAAAAGCI4AAAAADBEcAAAAABg6H8BBMzCQAJ/5CoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 864x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(\"domain_labels\", y=\"score\", hue='mother_college', data=test_scores,\n",
    "            kind=\"box\", color='white', fliersize=0)\n",
    "fig = plt.gcf()\n",
    "fig.set_size_inches(12, 5)\n",
    "fig.axes[0].set_xlabel('');"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}