Séance 4 - Statistiques avancées.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Python pour les SHS - Des statistiques descriptives aux traitements avancés"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Séance pratique plutôt orientée sciences sociales. La fin aborde des usages plus techniques, non abordés pendant le cours (manque de temps)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Objectif \n",
    "\n",
    "De nombreuses enquêtes par questionnaires sont disponibles. Dans le cas d'une collaboration sur la perception de la science en France, il est intéressant de réanalyser des données déjà existantes. \n",
    "\n",
    "Cela amène à réaliser les différentes opérations d'une analyse de données :\n",
    "\n",
    "- Charger les données et les nettoyer\n",
    "- Recoder et sélectionner\n",
    "- Décrire les données et les visualiser\n",
    "- Utiliser des modèles plus avancées\n",
    "\n",
    "Données sur le Github : https://github.com/pyshs/Formation-unil/blob/main/Data/fr.cdsp.ddi.sav\n",
    "\n",
    "- Nous allons faire certaines étapes, vous pouvez aller plus loin en intégrant plus de variables\n",
    "- Dans la vraie vie il y a tout une phase d'essais, d'erreurs, de suppression, etc.\n",
    "- L'organisation de ce document ne suit pas exactement un document final, avec l'idée qu'il vaut mieux au fur et à mesure regrouper les blocs déjà faits\n",
    "- Un processus de formalisation (des fonctions, etc.) se fait progressivement des usages\n",
    "- Comme toujours il y a plusieurs manières de faire les mêmes traitements"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Exploration de données"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Première étape, charger et parcourir\n",
    "\n",
    "Attention, on utilise beaucoup de bibliothèques différentes : pensez à les installer avant de les importer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Charger des données SPSS de l'enquête 2011 avec pyreadstat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 303,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(1027, 188)"
      ]
     },
     "execution_count": 303,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pyreadstat\n",
    "data2011, meta2011 = pyreadstat.read_sav('./cdsp_science2011/fr.cdsp.ddi.sav')\n",
    "data2011.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Quelques remarques : \n",
    "\n",
    "- ce sont des données pondérées -> qu'en faire ?\n",
    "- beaucoup de questions -> il faudra en faire un choix ET l'étape de prise en main des données n'est pas liée à la programmation mais à la problématique"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Définir des fonctions utilitaires\n",
    "\n",
    "Travailler avec Python permet de facilement se doter des outils dont on a besoin pour des usages spécifiques...\n",
    "\n",
    "(actuellement, la limite est que beaucoup d'outils même générique ne sont pas encore écrit...)\n",
    "\n",
    "\n",
    "Par exemple, accéder rapidement :\n",
    "\n",
    "- aux questions\n",
    "- aux modalités"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 268,
   "metadata": {},
   "outputs": [],
   "source": [
    "def question(v,meta):\n",
    "    return meta2011.column_names_to_labels[v]\n",
    "\n",
    "def modalites(v,meta):\n",
    "    return meta.value_labels[meta.variable_to_label[v]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Mot \"Science\" : Opinions négatives'"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "question(\"m_q29_1\",meta2011)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{1.0: 'Primaire',\n",
       " 2.0: 'Intermédiaire',\n",
       " 3.0: 'Secondaire',\n",
       " 4.0: 'Supérieur',\n",
       " 5.0: 'Supérieur médical',\n",
       " 6.0: 'Supérieur scientifique',\n",
       " 888.0: 'Ne sait pas'}"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "modalites(\"nivet6\",meta2011)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Sélectionner les variables d'intérêt\n",
    "\n",
    "Toute analyse il faut identifier les variables d'intérêt. Cela passe par la lecture des données.\n",
    "\n",
    "Variables dépendantes :\n",
    "\n",
    "- 'q2': 'Intérêt pour la science'\n",
    "- 'q15': \"Apport de la science à l'Homme\"\n",
    "\n",
    "Variables indépendantes :\n",
    "\n",
    "-  'rs1': 'Sexe du répondant'\n",
    "-  'rs2': 'Age du répondant'\n",
    "-  'rs6': 'PCS du répondant',\n",
    "-  'pcsip': 'Profession du répondant',\n",
    "- 'rs14': 'Revenus mensuels du foyer du répondant',\n",
    "-  'degre': 'Degré de connaissance scientifique (recodage en 4 postes du questionnaire auto-administré)',\n",
    "-  'rs16a': 'Diplôme le plus élevé du répondant',\n",
    "- 'rs15': 'Intérêt pour la politique',\n",
    "- 'rs16': \"Auto-positionnement sur l'échelle gauche/droite\",\n",
    "- 'rs11a': 'Proximité avec un parti politique',\n",
    "- 'rs17': 'Religion du répondant',\n",
    "-  'nivet6': \"Niveau d'études du répondant (recodage en 6 postes)\",\n",
    "-  'q1_1': 'Confiance dans : La justice',\n",
    " 'q1_2': 'Confiance dans : Les médias',\n",
    " 'q1_3': 'Confiance dans : La police',\n",
    " 'q1_4': \"Confiance dans : L'Assemblée nationale\",\n",
    " 'q1_5': \"Confiance dans : L'administration\",\n",
    " 'q1_6': 'Confiance dans : La science',\n",
    " 'q1_7': 'Confiance dans : Les grandes entreprises',\n",
    " 'q1_8': 'Confiance dans : Les associations',\n",
    " 'q1_9': 'Confiance dans : Le Gouvernement',\n",
    " 'q1_10': 'Confiance dans : Les agences gouvernementales',\n",
    "\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Regarder les données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.0    534\n",
       "1.0    493\n",
       "Name: rs1, dtype: int64"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data2011[\"rs1\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pas très joli ... et pas de prise en compte de la pondération. \n",
    "\n",
    "On peut mettre un peu en forme \"à la main\"\n",
    "\n",
    "- mettre les labels\n",
    "- changer le nom de la variable\n",
    "- faire un tri à plat ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "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>sexe</th>\n",
       "      <th>Proportion</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Femme</th>\n",
       "      <td>534</td>\n",
       "      <td>52.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Homme</th>\n",
       "      <td>493</td>\n",
       "      <td>48.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       sexe  Proportion\n",
       "Femme   534        52.0\n",
       "Homme   493        48.0"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "\n",
    "# Recoder\n",
    "data2011[\"sexe\"] = data2011[\"rs1\"].replace(modalites(\"rs1\",meta2011))\n",
    "\n",
    "# Mettre en forme\n",
    "tableau = pd.DataFrame(data2011[\"sexe\"].value_counts())\n",
    "tableau[\"Proportion\"] = tableau[\"sexe\"].apply(lambda x : round(100*x/len(data2011),2))\n",
    "tableau"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ou bien utiliser une bibliothèque qui prend en compte la pondération ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>Effectif redressé</th>\n",
       "      <th>Pourcentage (%)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sexe</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Femme</th>\n",
       "      <td>538.1</td>\n",
       "      <td>52.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Homme</th>\n",
       "      <td>488.9</td>\n",
       "      <td>47.6</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Effectif redressé  Pourcentage (%)\n",
       "sexe                                     \n",
       "Femme              538.1             52.4\n",
       "Homme              488.9             47.6"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pyshs\n",
    "\n",
    "tableau = pyshs.tri_a_plat(data2011,\"sexe\",\"poids1\")\n",
    "tableau"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualiser"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x12f35c950>"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "tableau[\"Effectif redressé\"].plot(kind=\"pie\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Continuer avec les différentes variables\n",
    "\n",
    "- recoder les différentes variables d'inérêt pour être sur\n",
    "- recoder les ages\n",
    "- créer un indicateur de confiance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 269,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Recodage des variables\n",
    "\n",
    "data2011[\"interetscience\"] = data2011['q2'].replace({1:\"Oui\",2:\"Oui\",3:\"Non\",4:\"Non\"})\n",
    "data2011[\"apportscience\"] = data2011['q15'].replace(modalites('q15',meta2011))\n",
    "data2011[\"politique\"] = data2011[\"rs16\"].replace({1.0: 'Extrême gauche',\n",
    " 2.0: 'Gauche',\n",
    " 3.0: 'Gauche',\n",
    " 4.0: '.Centre',\n",
    " 5.0: 'Droite',\n",
    " 6.0: 'Droite',\n",
    " 7.0: 'Extrême droite',\n",
    " 888.0: 'Ne sait pas'})\n",
    "\n",
    "data2011[\"sexe\"] = data2011[\"rs1\"].replace(modalites(\"rs1\",meta2011))\n",
    "data2011[\"age_reco\"] = pd.cut(data2011['rs2'],[0,35,45,55,65,100],\n",
    "                labels = [\"[18-35[\",\"[35-45[\",\"[45-55[\",\"[55-65[\",\"[65-75]\"])\n",
    "data2011[\"pcs\"] = data2011[\"pcsip\"].replace(modalites(\"pcsip\",meta2011))\n",
    "data2011[\"etude\"] = data2011[\"nivet6\"].replace(modalites(\"nivet6\",meta2011))\n",
    "data2011[\"politique\"] = data2011[\"rs16\"].replace({1.0: 'Extrême gauche',\n",
    " 2.0: 'Gauche',\n",
    " 3.0: 'Gauche',\n",
    " 4.0: '.Centre',\n",
    " 5.0: 'Droite',\n",
    " 6.0: 'Droite',\n",
    " 7.0: 'Extrême droite',\n",
    " 888.0: 'Ne sait pas'})\n",
    "\n",
    "data2011[\"confiance\"] = data2011[[\"q1_{}\".format(i) for i in range(1,11)]]\\\n",
    "                         .replace({1:1,2:1,3:0,4:0}).fillna(0).sum(axis=1)/10\n",
    "data2011[\"confiance_quar\"] = pd.qcut(data2011[\"confiance\"],[0,0.25,0.5,0.75,1],[\"Q1\",\"Q2\",\"Q3\",\"Q4\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "C'est un processus continu, et au fil de l'analyse il est fréquent de modifier"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Commencer à analyser"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "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>Effectif redressé</th>\n",
       "      <th>Pourcentage (%)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>interetscience</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Non</th>\n",
       "      <td>462.1</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Oui</th>\n",
       "      <td>564.0</td>\n",
       "      <td>55.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Effectif redressé  Pourcentage (%)\n",
       "interetscience                                    \n",
       "Non                         462.1             45.0\n",
       "Oui                         564.0             55.0"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pyshs.tri_a_plat(data2011,\"interetscience\",\"poids1\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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>Effectif redressé</th>\n",
       "      <th>Pourcentage (%)</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>apportscience</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>A peu près autant de bien que de mal</th>\n",
       "      <td>531.7</td>\n",
       "      <td>52.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Plus de bien que de mal</th>\n",
       "      <td>420.7</td>\n",
       "      <td>41.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Plus de mal que de bien</th>\n",
       "      <td>60.4</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      Effectif redressé  Pourcentage (%)\n",
       "apportscience                                                           \n",
       "A peu près autant de bien que de mal              531.7             52.5\n",
       "Plus de bien que de mal                           420.7             41.5\n",
       "Plus de mal que de bien                            60.4              6.0"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tableau = pyshs.tri_a_plat(data2011,\"apportscience\",\"poids1\")\n",
    "tableau"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x131f1c990>"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAi8AAAD4CAYAAAAkcAb6AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAcNUlEQVR4nO3deZxkZX3v8c+XGRwQyLAMchENo4gLIqAsweDCkmtCMMFlEjX4ihoiblGjIV688V4xMbkkmmiMEUWIGENAQFECkeUihMWNmWFgBhCNMMSgCWpgBFmE4Zc/6mkoeqp7apiuLk/zeb9e/eqzPPWc31NV0N95zjlVqSokSZK6YpNxFyBJkrQhDC+SJKlTDC+SJKlTDC+SJKlTDC+SJKlT5o+7AGmuW7RoUS1evHjcZUhSpyxbtuyHVbX9oH2GF2nEFi9ezNKlS8ddhiR1SpKbp9rnaSNJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQphhdJktQp88ddgDTXrbxlDYuPOXfcZTxo9XGHjbsESdoozrxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbxIkqROMbzMoiRrk6xIsirJGUke27bfOYJjrU6yaKb7neZ4Mz6GDTj2sUmO3tj2SR6f5MyZrU6SNNMML7Pr7qraq6p2B34KvHHcBekhVfW9qloy7jokSdMzvIzPZcBT+jckOTDJOX3rH03y2rZ8XJLrklyT5IOTO0uyXZILklyb5EQgffteneQbbdbnE0nmDXj86iT/r7VZmuQ5Sc5P8p0kb2xttkxyUZLlSVYmOXx9g0zyR0m+leTyJKdOzHgkuSTJPm15UZLVbXlekg8kubKN9Q3r6xd4Wt/2XZKcl2RZksuSPH2K0vZM8tUk307y+vbYxUlWTVdHe40uSXJmkm8mOSVJpjiGJGkE5o+7gEejJPOBQ4Hzhmy/HfBS4OlVVUm2HtDsvcDlVfXHSQ4DjmyPfQbwCuCAqrovyceAI4C/H9DHv1XVXkk+BJwMHABsBqwCPg7cA7y0qn7cTkl9LcnZVVVT1L038EpgL3rvteXAsvUM90hgTVXtm2QBcEWSC6rqpiH7PQF4Y1V9O8kvAB8DDh5wnD2A/YEtgKuSnDtMHW3fs4FnAt8DrmjP0+XrGZckaYYYXmbX5klWtOXLgJOGfNwaesHhpDYzc86ANi8AXgZQVecmua1tPwTYG7iyTRBsDtw6xXHObr9XAltW1R3AHUnubYHpJ8CfJXkB8ACwE7AD8B9T9Pd84KyqugsgydlTtOv3ImCPJBOnbxYCuwI39bUZ2G+SLYFfBM7omwxZMMVxvlhVdwN3J7kY2A9Y0bd/qjp+Cnyjqv69HXMFsJhJ4SXJUcBRAPN+bvshhi1JGpbhZXbdXVV7TbP/fh5+Km8zgKq6P8l+9ILIEuD3GDybMEiAT1fVu4doe2/7/UDf8sT6fHozNtsDe7dZnNUTNT4C/WPt7yPAW6vq/EfQ5ybA7et5jidMni2avD6wjiQH8vDnZi0D/juqqhPozQKxYMddB85MSZIeGa95+dlyM7BbkgVtpuMQeHBGYWFV/TPwDmDPAY+9FPit1v5QYJu2/SJgSZLHtX3bJtn5Eda3ELi1BZeDgPX1cynwkiSbJ9kK+LW+favpzQhBL5BNOB94U5JNW71PTbLFMP1W1Y+Bm5L8Rntskgx6rgAOT7JZOyV3IHDlpP3D1CFJGgNnXn6GVNV3k5xO7xqTm4Cr2q6tgC8m2YzejMA7Bzz8fcCpSa4FvgL8W+vzuiTvAS5IsglwH/AWekFpQ50C/FOSlcBS4JvrGc/yJJ8FrqZ3qqo/IHwQOL2dXum/3uREeqdhlrcLYX8AvGQD+j0COL6NeVPgtNZusmuAi4FFwJ9U1feSLN6QOiRJ45EprrWUZlySY4E7q2qdu6XmsgU77lo7vubD4y7jQauPO2zcJUjSeiVZVlX7DNrnaSNJktQpnjbSrKmqY8ddgySp+5x5kSRJnWJ4kSRJnWJ4kSRJnWJ4kSRJnWJ4kSRJnWJ4kSRJnWJ4kSRJnWJ4kSRJnWJ4kSRJnWJ4kSRJnWJ4kSRJneJ3G0kj9qydFrLUb3KWpBnjzIskSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeoUw4skSeqUocJLkh2SnJTkS219tyRHjrY0SZKkdQ0783IycD7w+Lb+LeD3R1GQJEnSdIYNL4uq6nTgAYCquh9YO7KqJEmSpjBsePlJku2AAkiyP7BmZFVJkiRNYf6Q7d4JnA3skuQKYHtgyciqkiRJmsJQ4aWqlid5IfA0IMANVXXfSCuTJEkaYNi7jd4CbFlV11bVKmDLJG8ebWmSJEnrGvaal9dX1e0TK1V1G/D60ZQkSZI0tWHDy7wkmVhJMg94zGhKkiRJmtqwF+yeB3w2ySfa+hvaNkmSpFk1bHj5X/QCy5va+oXAiSOpSJIkaRrD3m30AHB8+5EkSRqbocJLkgOAY4Gd22MCVFU9eXSlSZIkrWvY00YnAe8AluHXAkiSpDEaNrysqaovjbQSaY5aecsaFh9z7rjLkDQLVh932LhLeFQYNrxcnOQDwOeBeyc2VtXykVQlSZI0hWHDyy+03/v0bSvg4JktR5IkaXrD3m100KgLkSRJGsaw3220Q5KTknypre+W5MjRliZJkrSuYb8e4GTgfODxbf1bwO+PoiBJkqTpDBteFlXV6cADAFV1P94yLUmSxmDY8PKTJNvRu0iXJPsDa0ZWlSRJ0hSGvdvoncDZwC5JrgC2B5aMrCpJkqQpDHu30fIkLwSeRu+rAW6oqvtGWpkkSdIA04aXJAdX1ZeTvGzSrqcmoao+P8LaJEmS1rG+mZcXAl8Gfm3AvqL3ibuSJEmzZtrwUlXvbb9fNzvlSJIkTW/YD6n7syRb961vk+T9oytLkiRpsGFvlT60qm6fWKmq24BfHU1JkiRJUxs2vMxLsmBiJcnmwIJp2muAJGuTrEiyKskZSR7btt85gmOtTrJoY9sneWOS357Z6kYnybFJjp7F452cxI8NkKRZNOznvJwCXJTkU239dcCnR1PSnHZ3Ve0FkOQU4I3AX423pOlV1cfHXYMkSf2Gmnmpqj8H3g88o/38SVX9xSgLexS4DHhK/4YkByY5p2/9o0le25aPS3JdkmuSfHByZ0m2S3JBkmuTnEjv83gm9r06yTfarM8nksyboqZ3JVnZ2j6lPfbBmYwkuyQ5L8myJJcleXrbfnKSjyT5SpIbp5qJSPJHSb6V5PIkp/b1e0mSfdryoiSr2/K8JB9IcmUb9xvW1y+9zyKa2D6w3kmPPTbJp9v+m5O8LMlftOfhvCSbtnb/t9WxKskJSTK5L0nS7Bj2gt0tgAuq6mjgk8CCif+pa8MlmQ8cCqwcsv12wEuBZ1bVHvSC5GTvBS6vqmcCZwE/3x77DOAVwAFt1mctcMQUh1pTVc8CPgp8eMD+E4C3VtXewNHAx/r27Qg8D3gxcNyAMewNvBLYi971UvtON+bmyFbTvq3965M8aQP6na7efrsABwO/DvwDcHF7Hu4GDmttPlpV+1bV7sDmbZxTSnJUkqVJlq69y2/SkKSZNOxpo0uB5yfZBjgPWErvD+JUfwQ12OZJVrTly4CThnzcGuAe4KQ2M3POgDYvAF4GUFXnJrmtbT8E2Bu4sk0WbA7cOsVxTu37/aH+HUm2BH4ROKNv0qH/uqcvVNUDwHVJdhjQ9/OBs6rqrtbf2VPU0O9FwB59MzkLgV2Bm9bX7xD19vtSVd2XZCUwj957HHrhcnFbPijJu4DHAtsC1wL/NFXhVXUCvfDEgh13rSHGKkka0rDhJVV1V5IjgeOr6i/6/ghreA9e8zKF+3n4bNhm0PsW7yT70QsiS4DfozdTMIwAn66qdw/RtqZYptV1+zT13zvpmBuif9ybTernrVV1/gb2B+uvt9+9AFX1QJL7qmpi7A8A85NsRm/WZp+q+m6SYyfVKUmaRcPebZQkz6U303Ju2zbVdRN65G4GdkuyIL3P1TkEHpxFWFhV/wy8A9hzwGMvBX6rtT8U2KZtvwhYkuRxbd+2SXae4viv6Pv91f4dVfVj4KYkv9H6SZJBdUzlUuAlSTZPshUP/9Tm1fRmh+DhX/h5PvCmvutOntpOYa633xmot99EUPlhey28u0iSxmjYmZe3A++mNz1/bZInAxePrqxHp/av+tOBVfROjVzVdm0FfLHNAITet3xP9j7g1CTXAl8B/q31eV2S9wAXJNkEuA94C72gNNk2Sa6hNxPxqgH7jwCOb/1tCpwGXD3k2JYn+WxrfytwZd/uDwKnJzmKh8IxwIn0TtssbxfI/gB4yQb0+4jrnXSM25N8kt7r8h+TjiFJmmV5aIZ8mkbJb1TVGevbJg2rnXq5s6rWuXNqrlmw466142sGXf8saa5Zfdxh62+koSRZVlX7DNo37GmjQddLDHMNhSRJ0oya9rRRu3biV4Gdknykb9fP0bvIUnpEqurYcdcgSeqm9V3z8j16t0X/OrCsb/sd9C4clSRJmlXThpequjrJKuCXq8qvA5AkSWO33mteqmot8MQkj5mFeiRJkqY17K3SNwFXtE8v/cnExqr6mf5SQUmSNPcMG16+0342ofeZI5IkSWMxVHipqvfBg5/0SlXdOcqiJEmSpjLst0rvnuQqel9Gd22SZUmeOdrSJEmS1jXsh9SdALyzqnauqp2BPwA+ObqyJEmSBhs2vGxRVQ9+l1FVXQJM/oI8SZKkkRv2gt0bk/wf4DNt/dXAjaMpSZIkaWrDhpffofetxZ9v65e1bZLW41k7LWSpX9YmSTNm2LuNbgPelmQh8EBV3THasiRJkgYb9m6jfZOsBK4GVia5Osneoy1NkiRpXcOeNjoJeHNVXQaQ5HnAp4A9RlWYJEnSIMPebbR2IrgAVNXlwP2jKUmSJGlqw868/EuSTwCnAgW8ArgkyXMAqmr5iOqTJEl6mGHDy57t93snbX82vTBz8IxVJEmSNI1h7zY6aNSFSJIkDWPYu422S/KRJMvb9xr9dZLtRl2cJEnSZMNesHsa8APg5cCStvzZURUlSZI0lWGvedmxqv6kb/39SV4xioIkSZKmM+zMywVJXplkk/bzm8D5oyxMkiRpkGHDy+uBfwTubT+nAW9IckeSH4+qOEmSpMmGvdtoqyTbArsCm/Vt/5dRFSZJkjTIUOElye8CbweeAKwA9ge+AhwyutIkSZLWNexpo7cD+wI3t898eTawZmRVSZIkTWHY8HJPVd0DkGRBVX0TeNroypIkSRps2Ful/z3J1sAXgAuT3AbcPLqyJEmSBhv2gt2XtsVjk1wMLATOG1lVkiRJUxh25uVB3mEkSZLGadhrXiRJkn4mGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnGF4kSVKnzB93AdJct/KWNSw+5txxlyFJs2r1cYeNrG9nXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqcYXiRJUqc86sNLkqcmOXzcdUiSpOGMNLwkeUmSSvL0UR5nY1TVt4BnJ3npKI+TZOskb97IPl6b5PFDtFucZNUG9Dtl+yQnJtltQ+ocpySXJNlnFo+3Osmi2TqeJGn0My+vAi5vv3/mJJkHUFXHVtVZIz7c1sBGhRfgtcB6w8tMqqrfrarrZvOYkiRNZ2ThJcmWwPOAI4FXTtFmcZJvJjklyfVJzkzy2LZv7yT/kmRZkvOT7Ni2P/gv6ySLkqwe0O+BSS5Ncm6SG5J8PMkmbd+dSf4yydXAc5O8Osk3klyd5BNJ5rWfk5OsSrIyyTsGHOPXknw9yVVJ/n+SHdr2Y5Mc3dduVZLFwHHALklWJPlAki2TXJRkeTvG4X3PyfVJPpnk2iQXJNk8yRJgH+CU1sfmk+rZu43hauAtfdvnteNdmeSaJG+Y4iWbP8Xr0P98vyjJV1vNZ7TXeGL24X19Y1lnpq2N4bTW/1ntuZvo986+dkuSnNyWt0/yuVb7lUkOWF+/wOZ9+wbWO+nxlyT5UJKlrY99k3w+ybeTvL+v3Rfae/HaJEdN8RxKkmbBKGdeDgfOa6dlfpRk7ynaPQ34WFU9A/gx8OYkmwJ/Ayypqr2BvwP+dAOPvx/wVmA3YBfgZW37FsDXq2pP4Ef0gtUBbR3g1cBewE5VtXtVPQv41ID+Lwf2r6pnA6cB71pPPccA36mqvarqD4F7gJdW1XOAg4C/TJLWdlfgb6vqmcDtwMur6kxgKXBE6+PuSf1/Cnhr3zgmHAmsqap9gX2B1yd50oD61nkd+ne2UyPvAX6p1bwUeGdfkx+27ccDR7OuNwF3tf7fC0z1fuj318CHWu0vB04ctt8h6u3306raB/g48EV64W934LVJtmttfqe9F/cB3ta3faAkR7VAtHTtXWuGGKokaVjzR9j3q+j98YHeH/dXAcsGtPtuVV3Rlv8BeBtwHr0/Hhe2v+fzgO9v4PG/UVU3AiQ5ld4s0JnAWuBzrc0hwDP6jrMl8F3gbODJSf4GOBe4YED/TwA+22aEHgPctIH1BfizJC8AHgB2AnZo+26qqhVteRmweNqOkq2Bravq0rbpM8ChbflFwB5t5gZgIb1wNLneQa/DB/v2708vCF7RnqvHAF/t2//5vnpfxrpeAHwEoKquSXLNdGNqfgnY7aFMx88l2bKq7uxrM1W/66u339nt90rg2qr6PkCSG4En0gu5b8tD10U9kd5z+KOpCq+qE4ATABbsuGsNMVZJ0pBGEl6SbAscDDwrSdELH5XkD6tq8v/IB62H3h+R5w7o/n4emjHabJoypjrOPVW1dqJU4IyqOmbAGPYEfhl4I/CbwO9MavI3wF9V1dlJDgSOHVDfdDUeAWwP7F1V96V3+mui7b197dbSdyrkEQi9GZnz19Nufa9LgAuraqrrlyZqXsuGv6/6j9X/fG1Cb3brng3sD9Zfb7+J2h/g4c/9A/ROpx1IL0g9t6ruSnIJ07/3JEkjNKrTRkuAz1TVzlW1uKqeSO9f+s8f0Pbnk0yElN+idzrmBmD7ie1JNk3yzNZmNQ+dcljC1PZL8qT0rnV5Ret3souAlyd5XDvOdu2ak0XAJlX1OXqnHp4z4LELgVva8mv6tq+eaJ/kOcDEKZo7gK0mPf7WFlwOAnaeZiwTJvcBQFXdDtye5Hlt0xF9u88H3tROxU3cGr7FgL4HvQ79vgYckOQprZ8tkjx1iJonXNr6JcnuwB59+/4zyTPaa9V/19cF9E790R631wb0u7H19lsI3NaCy9PpzepIksZkVOHlVcDku3c+x+C7jm4A3pLkemAb4Piq+im9YPLn7QLUFcAvtvYfpPfH+CpgultUrwQ+ClxPLzitczdRu4vmPcAF7XTDBcD/oHcK55IkK+idQnn3gP6PBc5Isgz44aRxbpvkWuD3gG+1Y/2I3imMVUk+AJwC7JNkJfDbwDenGcuEk4GPZ8AFu8DrgL9tNadv+4nAdcDy9G6H/gSDZ0bWeR36d1bVD+jd7XRqe66+CmzILfDHA1u2/v+Yh59CPAY4B/gKDz89+DZ6z9E1Sa6jNws2VL8zUG+/8+jNwFxP78Lrrz3CfiRJMyDrnsWZxYP37sI5p6p2n+F+DwSOrqoXz2S/mjnt1MvRVbV03LWM2oIdd60dX/PhcZchSbNq9XGHbdTjkyxrN1Os41H/CbuSJKlbRnm30XpV1Wp6dxXNdL+XAJfMdL+aOVV14LhrkCR1kzMvkiSpUwwvkiSpUwwvkiSpUwwvkiSpUwwvkiSpUwwvkiSpUwwvkiSpUwwvkiSpUwwvkiSpUwwvkiSpU8b69QDSo8GzdlrI0o38gjJJ0kOceZEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ2Sqhp3DdKcluQO4IZx1zGLFgE/HHcRs8jxzm2Od3x2rqrtB+2YP9uVSI9CN1TVPuMuYrYkWep45y7HO7d1ZbyeNpIkSZ1ieJEkSZ1ieJFG74RxFzDLHO/c5njntk6M1wt2JUlSpzjzIkmSOsXwIkmSOsXwIo1Qkl9JckOSf01yzLjrmQlJ/i7JrUlW9W3bNsmFSb7dfm/TtifJR9r4r0nynPFVvuGSPDHJxUmuS3Jtkre37XN1vJsl+UaSq9t439e2PynJ19u4PpvkMW37grb+r23/4nHW/0glmZfkqiTntPU5O94kq5OsTLIiydK2rXPvZ8OLNCJJ5gF/CxwK7Aa8Kslu461qRpwM/MqkbccAF1XVrsBFbR16Y9+1/RwFHD9LNc6U+4E/qKrdgP2Bt7TXcK6O917g4KraE9gL+JUk+wN/Dnyoqp4C3AYc2dofCdzWtn+oteuitwPX963P9fEeVFV79X2eS+fez4YXaXT2A/61qm6sqp8CpwGHj7mmjVZVlwL/NWnz4cCn2/KngZf0bf/76vkasHWSHWen0o1XVd+vquVt+Q56f+B2Yu6Ot6rqzra6afsp4GDgzLZ98ngnnoczgUOSZJbKnRFJngAcBpzY1sMcHu8UOvd+NrxIo7MT8N2+9X9v2+aiHarq+235P4Ad2vKceQ7aKYJnA19nDo+3nUJZAdwKXAh8B7i9qu5vTfrH9OB42/41wHazW/FG+zDwLuCBtr4dc3u8BVyQZFmSo9q2zr2f/XoASTOqqirJnPoMhiRbAp8Dfr+qftz/j+25Nt6qWgvslWRr4Czg6WMuaWSSvBi4taqWJTlw3PXMkudV1S1JHgdcmOSb/Tu78n525kUanVuAJ/atP6Ftm4v+c2I6uf2+tW3v/HOQZFN6weWUqvp82zxnxzuhqm4HLgaeS+90wcQ/dvvH9OB42/6FwI9mudSNcQDw60lW0zutezDw18zd8VJVt7Tft9ILp/vRwfez4UUanSuBXdudC48BXgmcPeaaRuVs4DVt+TXAF/u2/3a7a2F/YE3f9PTPvHY9w0nA9VX1V3275up4t28zLiTZHPif9K7zuRhY0ppNHu/E87AE+HJ16JNPq+rdVfWEqlpM77/PL1fVEczR8SbZIslWE8vAi4BVdPD97CfsSiOU5FfpnVOfB/xdVf3pmEvaaElOBQ4EFgH/CbwX+AJwOvDzwM3Ab1bVf7U//h+ld3fSXcDrqmrpOOp+JJI8D7gMWMlD10T8b3rXvczF8e5B74LNefT+cXt6Vf1xkifTm5nYFrgKeHVV3ZtkM+Az9K4F+i/glVV143iq3zjttNHRVfXiuTreNq6z2up84B+r6k+TbEfH3s+GF0mS1CmeNpIkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ1ieJEkSZ3y37+vPmmnvvaGAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "tableau[\"Effectif redressé\"].plot(kind=\"barh\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Des graphiques un peu plus complexes.\n",
    "\n",
    "Par exemple la confiance"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 10 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in [\"q1_{}\".format(i) for i in range(1,11)]:\n",
    "    data2011[i] = data2011[i].replace({1.0: '1 - Très confiance',\n",
    " 2.0: '2 - Plutôt confiance',\n",
    " 3.0: '3 - Plutôt pas confiance',\n",
    " 4.0: '4 - Pas du tout confiance',\n",
    " 888.0: 'Ne sait pas'})\n",
    "\n",
    "labels = [\"La justice\",\"Les médias\",\"La police\",\"L'Assemblée nationale\",\"L'administration\",\n",
    "\"La science\",\"Les grandes entreprises\",\"Les associations\",\"Le Gouvernement\",\"Les agences gouvernementales\"]\n",
    "\n",
    "fix,ax = plt.subplots(5,2,figsize=(10,10),sharex=True)\n",
    "for i in zip(range(1,11),ax.flat):\n",
    "    pyshs.tri_a_plat(data2011,\"q1_{}\".format(i[0]),\"poids1\")[\"Pourcentage (%)\"].plot(ax=i[1],\n",
    "                                        kind=\"bar\",rot=45,color=[\"green\",\"yellow\",\"orange\",\"red\"],alpha=0.5)\n",
    "    i[1].set_title(labels[i[0]-1])\n",
    "    i[1].set_xlabel(\"\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### L'analyse bivariée\n",
    "\n",
    "L'étape souvent ensuite est l'analyse bivariée, qui passe (entre autre) par des tableaux"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "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>sexe</th>\n",
       "      <th>Femme</th>\n",
       "      <th>Homme</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>interetscience</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Non</th>\n",
       "      <td>270</td>\n",
       "      <td>196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Oui</th>\n",
       "      <td>264</td>\n",
       "      <td>296</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "sexe            Femme  Homme\n",
       "interetscience              \n",
       "Non               270    196\n",
       "Oui               264    296"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.crosstab(data2011[\"interetscience\"],data2011[\"sexe\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "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>interetscience</th>\n",
       "      <th>Non</th>\n",
       "      <th>Oui</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sexe</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Femme</th>\n",
       "      <td>272.2 (50.6%)</td>\n",
       "      <td>265.8 (49.4%)</td>\n",
       "      <td>538.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Homme</th>\n",
       "      <td>189.8 (38.9%)</td>\n",
       "      <td>298.1 (61.1%)</td>\n",
       "      <td>487.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>462.1</td>\n",
       "      <td>564</td>\n",
       "      <td>1026.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "interetscience            Non            Oui     All\n",
       "sexe                                                \n",
       "Femme           272.2 (50.6%)  265.8 (49.4%)   538.1\n",
       "Homme           189.8 (38.9%)  298.1 (61.1%)   487.9\n",
       "All                     462.1            564  1026.0"
      ]
     },
     "execution_count": 81,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pyshs.tableau_croise(data2011,\"sexe\",\"interetscience\",\"poids1\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "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>apportscience</th>\n",
       "      <th>A peu près autant de bien que de mal</th>\n",
       "      <th>Plus de bien que de mal</th>\n",
       "      <th>Plus de mal que de bien</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>sexe</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Femme</th>\n",
       "      <td>301.0 (57.1%)</td>\n",
       "      <td>196.3 (37.2%)</td>\n",
       "      <td>30.1 (5.7%)</td>\n",
       "      <td>527.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Homme</th>\n",
       "      <td>230.7 (47.5%)</td>\n",
       "      <td>224.4 (46.2%)</td>\n",
       "      <td>30.3 (6.2%)</td>\n",
       "      <td>485.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>531.7</td>\n",
       "      <td>420.7</td>\n",
       "      <td>60.4</td>\n",
       "      <td>1012.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "apportscience A peu près autant de bien que de mal Plus de bien que de mal  \\\n",
       "sexe                                                                         \n",
       "Femme                                301.0 (57.1%)           196.3 (37.2%)   \n",
       "Homme                                230.7 (47.5%)           224.4 (46.2%)   \n",
       "All                                          531.7                   420.7   \n",
       "\n",
       "apportscience Plus de mal que de bien     All  \n",
       "sexe                                           \n",
       "Femme                     30.1 (5.7%)   527.4  \n",
       "Homme                     30.3 (6.2%)   485.4  \n",
       "All                              60.4  1012.8  "
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pyshs.tableau_croise(data2011,\"sexe\",\"apportscience\",\"poids1\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Faire un test d'hypothèse\n",
    "\n",
    "Identifier une des bibliothèques qui le fait, par exemple un chi2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "metadata": {},
   "outputs": [],
   "source": [
    "from scipy.stats import chi2_contingency"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 270,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8.553196646789049,\n",
       " 0.013889830526662587,\n",
       " 2,\n",
       " array([[277.25962488, 212.60019743,  34.14017769],\n",
       "        [258.74037512, 198.39980257,  31.85982231]]))"
      ]
     },
     "execution_count": 270,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chi2_contingency(pd.crosstab(data2011[\"sexe\"],data2011[\"apportscience\"]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Savoir quel test faire etc. relève pour le coup des attentes de chaque domaine ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Intégrer les variables dans un modèle\n",
    "\n",
    "Il existe une diversité de modèles qui permettent d'évaluer les effets conjugués de différentes variables dans la détermination d'une variable dépendante\n",
    "\n",
    "Chaque modèle renvoie à une littérature.\n",
    "\n",
    "Dans notre cas ici, nous avons comme souvent en SHS des variables catégorielles, on va utiliser une régresssion logistique de la bibliothèque statsmodel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Construire le modèle"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Dichotomiser la variable\n",
    "data2011[\"apportsciencepositif\"] = data2011[\"apportscience\"].apply(lambda x : 1 if x == \"Plus de bien que de mal\" else 0)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Chargement de la bibliothèque\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.formula.api as smf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 205,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Définition de la formule\n",
    "f = \"apportsciencepositif ~ sexe + age_reco + confiance_quar + etude + politique\"\n",
    "\n",
    "# Création du modèle \n",
    "modele = smf.glm(formula=f, data=data2011, family=sm.families.Binomial(), freq_weights=data2011[\"poids1\"])\n",
    "regression = modele.fit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Regarder les résultats"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 206,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<table class=\"simpletable\">\n",
       "<caption>Generalized Linear Model Regression Results</caption>\n",
       "<tr>\n",
       "  <th>Dep. Variable:</th>   <td>apportsciencepositif</td> <th>  No. Observations:  </th>  <td>   856</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model:</th>                    <td>GLM</td>         <th>  Df Residuals:      </th> <td>  835.56</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Model Family:</th>          <td>Binomial</td>       <th>  Df Model:          </th>  <td>    17</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Link Function:</th>           <td>logit</td>        <th>  Scale:             </th> <td>  1.0000</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Method:</th>                  <td>IRLS</td>         <th>  Log-Likelihood:    </th> <td> -526.65</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Date:</th>              <td>Wed, 28 Apr 2021</td>   <th>  Deviance:          </th> <td>  1053.3</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Time:</th>                  <td>22:28:51</td>       <th>  Pearson chi2:      </th>  <td>  852.</td> \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>No. Iterations:</th>            <td>4</td>          <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Covariance Type:</th>       <td>nonrobust</td>      <th>                     </th>     <td> </td>   \n",
       "</tr>\n",
       "</table>\n",
       "<table class=\"simpletable\">\n",
       "<tr>\n",
       "                 <td></td>                    <th>coef</th>     <th>std err</th>      <th>z</th>      <th>P>|z|</th>  <th>[0.025</th>    <th>0.975]</th>  \n",
       "</tr>\n",
       "<tr>\n",
       "  <th>Intercept</th>                       <td>   -2.1838</td> <td>    0.279</td> <td>   -7.840</td> <td> 0.000</td> <td>   -2.730</td> <td>   -1.638</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>sexe[T.Homme]</th>                   <td>    0.4007</td> <td>    0.153</td> <td>    2.614</td> <td> 0.009</td> <td>    0.100</td> <td>    0.701</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_reco[T.[35-45[]</th>             <td>    0.3666</td> <td>    0.227</td> <td>    1.615</td> <td> 0.106</td> <td>   -0.078</td> <td>    0.812</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_reco[T.[45-55[]</th>             <td>    0.7876</td> <td>    0.247</td> <td>    3.189</td> <td> 0.001</td> <td>    0.303</td> <td>    1.272</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_reco[T.[55-65[]</th>             <td>    0.6302</td> <td>    0.229</td> <td>    2.753</td> <td> 0.006</td> <td>    0.182</td> <td>    1.079</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>age_reco[T.[65-75]]</th>             <td>    0.8471</td> <td>    0.237</td> <td>    3.574</td> <td> 0.000</td> <td>    0.383</td> <td>    1.312</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>confiance_quar[T.Q2]</th>            <td>    0.4565</td> <td>    0.196</td> <td>    2.324</td> <td> 0.020</td> <td>    0.071</td> <td>    0.842</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>confiance_quar[T.Q3]</th>            <td>    0.9698</td> <td>    0.233</td> <td>    4.158</td> <td> 0.000</td> <td>    0.513</td> <td>    1.427</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>confiance_quar[T.Q4]</th>            <td>    0.9809</td> <td>    0.214</td> <td>    4.584</td> <td> 0.000</td> <td>    0.562</td> <td>    1.400</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>etude[T.Primaire]</th>               <td>   -0.0753</td> <td>    0.244</td> <td>   -0.308</td> <td> 0.758</td> <td>   -0.554</td> <td>    0.403</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>etude[T.Secondaire]</th>             <td>    0.9073</td> <td>    0.225</td> <td>    4.024</td> <td> 0.000</td> <td>    0.465</td> <td>    1.349</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>etude[T.Supérieur]</th>              <td>    1.0396</td> <td>    0.211</td> <td>    4.932</td> <td> 0.000</td> <td>    0.626</td> <td>    1.453</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>etude[T.Supérieur médical]</th>      <td>    0.8142</td> <td>    0.490</td> <td>    1.663</td> <td> 0.096</td> <td>   -0.146</td> <td>    1.774</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>etude[T.Supérieur scientifique]</th> <td>    1.2163</td> <td>    0.267</td> <td>    4.555</td> <td> 0.000</td> <td>    0.693</td> <td>    1.740</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>politique[T.Droite]</th>             <td>    0.2892</td> <td>    0.210</td> <td>    1.375</td> <td> 0.169</td> <td>   -0.123</td> <td>    0.701</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>politique[T.Extrême droite]</th>     <td>   -0.5697</td> <td>    0.524</td> <td>   -1.088</td> <td> 0.277</td> <td>   -1.596</td> <td>    0.457</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>politique[T.Extrême gauche]</th>     <td>    0.6729</td> <td>    0.448</td> <td>    1.503</td> <td> 0.133</td> <td>   -0.205</td> <td>    1.550</td>\n",
       "</tr>\n",
       "<tr>\n",
       "  <th>politique[T.Gauche]</th>             <td>    0.1800</td> <td>    0.181</td> <td>    0.996</td> <td> 0.319</td> <td>   -0.174</td> <td>    0.534</td>\n",
       "</tr>\n",
       "</table>"
      ],
      "text/plain": [
       "<class 'statsmodels.iolib.summary.Summary'>\n",
       "\"\"\"\n",
       "                  Generalized Linear Model Regression Results                   \n",
       "================================================================================\n",
       "Dep. Variable:     apportsciencepositif   No. Observations:                  856\n",
       "Model:                              GLM   Df Residuals:                   835.56\n",
       "Model Family:                  Binomial   Df Model:                           17\n",
       "Link Function:                    logit   Scale:                          1.0000\n",
       "Method:                            IRLS   Log-Likelihood:                -526.65\n",
       "Date:                  Wed, 28 Apr 2021   Deviance:                       1053.3\n",
       "Time:                          22:28:51   Pearson chi2:                     852.\n",
       "No. Iterations:                       4                                         \n",
       "Covariance Type:              nonrobust                                         \n",
       "===================================================================================================\n",
       "                                      coef    std err          z      P>|z|      [0.025      0.975]\n",
       "---------------------------------------------------------------------------------------------------\n",
       "Intercept                          -2.1838      0.279     -7.840      0.000      -2.730      -1.638\n",
       "sexe[T.Homme]                       0.4007      0.153      2.614      0.009       0.100       0.701\n",
       "age_reco[T.[35-45[]                 0.3666      0.227      1.615      0.106      -0.078       0.812\n",
       "age_reco[T.[45-55[]                 0.7876      0.247      3.189      0.001       0.303       1.272\n",
       "age_reco[T.[55-65[]                 0.6302      0.229      2.753      0.006       0.182       1.079\n",
       "age_reco[T.[65-75]]                 0.8471      0.237      3.574      0.000       0.383       1.312\n",
       "confiance_quar[T.Q2]                0.4565      0.196      2.324      0.020       0.071       0.842\n",
       "confiance_quar[T.Q3]                0.9698      0.233      4.158      0.000       0.513       1.427\n",
       "confiance_quar[T.Q4]                0.9809      0.214      4.584      0.000       0.562       1.400\n",
       "etude[T.Primaire]                  -0.0753      0.244     -0.308      0.758      -0.554       0.403\n",
       "etude[T.Secondaire]                 0.9073      0.225      4.024      0.000       0.465       1.349\n",
       "etude[T.Supérieur]                  1.0396      0.211      4.932      0.000       0.626       1.453\n",
       "etude[T.Supérieur médical]          0.8142      0.490      1.663      0.096      -0.146       1.774\n",
       "etude[T.Supérieur scientifique]     1.2163      0.267      4.555      0.000       0.693       1.740\n",
       "politique[T.Droite]                 0.2892      0.210      1.375      0.169      -0.123       0.701\n",
       "politique[T.Extrême droite]        -0.5697      0.524     -1.088      0.277      -1.596       0.457\n",
       "politique[T.Extrême gauche]         0.6729      0.448      1.503      0.133      -0.205       1.550\n",
       "politique[T.Gauche]                 0.1800      0.181      0.996      0.319      -0.174       0.534\n",
       "===================================================================================================\n",
       "\"\"\""
      ]
     },
     "execution_count": 206,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "regression.summary()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Calculer une information manquante"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 214,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.1"
      ]
     },
     "execution_count": 214,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def pseudoR2(mod):\n",
    "    R = 1-mod.deviance/mod.null_deviance\n",
    "    return round(R,2)\n",
    "\n",
    "pseudoR2(regression)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Afficher un tableau bien mis en forme"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 208,
   "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>Odds Ratio</th>\n",
       "      <th>p</th>\n",
       "      <th>IC</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Variable</th>\n",
       "      <th>Modalité</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>.Intercept</th>\n",
       "      <th></th>\n",
       "      <td>0.11</td>\n",
       "      <td>0.0***</td>\n",
       "      <td>0.11 [0.07-0.19]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">Age</th>\n",
       "      <th>[18-35[</th>\n",
       "      <td>ref</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>[35-45[</th>\n",
       "      <td>1.44</td>\n",
       "      <td>0.106</td>\n",
       "      <td>1.44 [0.92-2.25]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>[45-55[</th>\n",
       "      <td>2.2</td>\n",
       "      <td>0.001**</td>\n",
       "      <td>2.20 [1.35-3.57]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>[55-65[</th>\n",
       "      <td>1.88</td>\n",
       "      <td>0.006**</td>\n",
       "      <td>1.88 [1.20-2.94]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>[65-75]</th>\n",
       "      <td>2.33</td>\n",
       "      <td>0.0***</td>\n",
       "      <td>2.33 [1.47-3.71]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">Confiance</th>\n",
       "      <th>Q1</th>\n",
       "      <td>ref</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Q2</th>\n",
       "      <td>1.58</td>\n",
       "      <td>0.02*</td>\n",
       "      <td>1.58 [1.07-2.32]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Q3</th>\n",
       "      <td>2.64</td>\n",
       "      <td>0.0***</td>\n",
       "      <td>2.64 [1.67-4.17]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Q4</th>\n",
       "      <td>2.67</td>\n",
       "      <td>0.0***</td>\n",
       "      <td>2.67 [1.75-4.06]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"6\" valign=\"top\">Niveau d'étude</th>\n",
       "      <th>Intermédiaire</th>\n",
       "      <td>ref</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Primaire</th>\n",
       "      <td>0.93</td>\n",
       "      <td>0.758</td>\n",
       "      <td>0.93 [0.57-1.50]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Secondaire</th>\n",
       "      <td>2.48</td>\n",
       "      <td>0.0***</td>\n",
       "      <td>2.48 [1.59-3.85]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Supérieur</th>\n",
       "      <td>2.83</td>\n",
       "      <td>0.0***</td>\n",
       "      <td>2.83 [1.87-4.27]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Supérieur médical</th>\n",
       "      <td>2.26</td>\n",
       "      <td>0.096</td>\n",
       "      <td>2.26 [0.86-5.90]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Supérieur scientifique</th>\n",
       "      <td>3.37</td>\n",
       "      <td>0.0***</td>\n",
       "      <td>3.37 [2.00-5.70]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">Politique</th>\n",
       "      <th>.Centre</th>\n",
       "      <td>ref</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Droite</th>\n",
       "      <td>1.34</td>\n",
       "      <td>0.169</td>\n",
       "      <td>1.34 [0.88-2.02]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Extrême droite</th>\n",
       "      <td>0.57</td>\n",
       "      <td>0.277</td>\n",
       "      <td>0.57 [0.20-1.58]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Extrême gauche</th>\n",
       "      <td>1.96</td>\n",
       "      <td>0.133</td>\n",
       "      <td>1.96 [0.81-4.71]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Gauche</th>\n",
       "      <td>1.2</td>\n",
       "      <td>0.319</td>\n",
       "      <td>1.20 [0.84-1.71]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">Sexe</th>\n",
       "      <th>Femme</th>\n",
       "      <td>ref</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Homme</th>\n",
       "      <td>1.49</td>\n",
       "      <td>0.009**</td>\n",
       "      <td>1.49 [1.11-2.02]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      Odds Ratio        p                IC\n",
       "Variable       Modalité                                                    \n",
       ".Intercept                                  0.11   0.0***  0.11 [0.07-0.19]\n",
       "Age            [18-35[                       ref                           \n",
       "               [35-45[                      1.44    0.106  1.44 [0.92-2.25]\n",
       "               [45-55[                       2.2  0.001**  2.20 [1.35-3.57]\n",
       "               [55-65[                      1.88  0.006**  1.88 [1.20-2.94]\n",
       "               [65-75]                      2.33   0.0***  2.33 [1.47-3.71]\n",
       "Confiance      Q1                            ref                           \n",
       "               Q2                           1.58    0.02*  1.58 [1.07-2.32]\n",
       "               Q3                           2.64   0.0***  2.64 [1.67-4.17]\n",
       "               Q4                           2.67   0.0***  2.67 [1.75-4.06]\n",
       "Niveau d'étude Intermédiaire                 ref                           \n",
       "               Primaire                     0.93    0.758  0.93 [0.57-1.50]\n",
       "               Secondaire                   2.48   0.0***  2.48 [1.59-3.85]\n",
       "               Supérieur                    2.83   0.0***  2.83 [1.87-4.27]\n",
       "               Supérieur médical            2.26    0.096  2.26 [0.86-5.90]\n",
       "               Supérieur scientifique       3.37   0.0***  3.37 [2.00-5.70]\n",
       "Politique      .Centre                       ref                           \n",
       "               Droite                       1.34    0.169  1.34 [0.88-2.02]\n",
       "               Extrême droite               0.57    0.277  0.57 [0.20-1.58]\n",
       "               Extrême gauche               1.96    0.133  1.96 [0.81-4.71]\n",
       "               Gauche                        1.2    0.319  1.20 [0.84-1.71]\n",
       "Sexe           Femme                         ref                           \n",
       "               Homme                        1.49  0.009**  1.49 [1.11-2.02]"
      ]
     },
     "execution_count": 208,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ind_var = {\"sexe\":\"Sexe\",\"age_reco\":\"Age\",\"confiance_quar\":\"Confiance\",\"etude\":\"Niveau d'étude\",\"politique\":\"Politique\"}\n",
    "pyshs.tableau_reg_logistique(regression,data2011,ind_var,sig=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Faire un graphique add hoc"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.lines.Line2D at 0x138729590>"
      ]
     },
     "execution_count": 271,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Construction des données\n",
    "u = regression.summary2()\n",
    "u = np.exp(u.tables[1][[\"Coef.\",\"[0.025\",\"0.975]\"]])\n",
    "u[\"low\"] = u[\"Coef.\"]-u[\"[0.025\"]\n",
    "u[\"up\"] = u[\"0.975]\"]-u[\"Coef.\"]\n",
    "\n",
    "# Création du graphique\n",
    "fig,ax = plt.subplots(figsize=(8,5))\n",
    "ax.errorbar( u[\"Coef.\"],u.index, xerr=u[[\"low\",\"up\"]].values.T,fmt='o', color='black',\n",
    "             ecolor='gray', elinewidth=5, capsize=10)\n",
    "plt.axvline(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Analyse factorielle sur la confiance"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Matrice de données numériques"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "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>q1_1</th>\n",
       "      <th>q1_2</th>\n",
       "      <th>q1_3</th>\n",
       "      <th>q1_4</th>\n",
       "      <th>q1_5</th>\n",
       "      <th>q1_6</th>\n",
       "      <th>q1_7</th>\n",
       "      <th>q1_8</th>\n",
       "      <th>q1_9</th>\n",
       "      <th>q1_10</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1022</th>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1023</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1024</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1025</th>\n",
       "      <td>2.5</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.5</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1026</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1027 rows × 10 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      q1_1  q1_2  q1_3  q1_4  q1_5  q1_6  q1_7  q1_8  q1_9  q1_10\n",
       "0      3.0   4.0   2.0   3.0   3.0   2.0   3.0   3.0   3.0    3.0\n",
       "1      2.0   3.0   1.0   2.0   2.0   2.0   3.0   3.0   2.0    2.0\n",
       "2      2.0   3.0   2.0   2.0   2.0   2.0   2.0   2.0   2.5    2.0\n",
       "3      2.0   3.0   2.0   3.0   2.0   2.0   3.0   2.0   4.0    2.0\n",
       "4      2.0   3.0   4.0   4.0   3.0   2.0   4.0   2.0   4.0    4.0\n",
       "...    ...   ...   ...   ...   ...   ...   ...   ...   ...    ...\n",
       "1022   2.0   2.0   2.0   2.0   2.0   1.0   2.0   2.0   3.0    3.0\n",
       "1023   2.0   3.0   2.0   2.0   3.0   2.0   2.0   2.0   3.0    2.0\n",
       "1024   2.0   3.0   2.0   2.0   2.0   1.0   3.0   3.0   3.0    3.0\n",
       "1025   2.5   2.0   2.0   2.0   2.5   2.0   2.0   2.0   2.0    2.0\n",
       "1026   2.0   3.0   2.0   3.0   3.0   2.0   3.0   2.0   3.0    3.0\n",
       "\n",
       "[1027 rows x 10 columns]"
      ]
     },
     "execution_count": 90,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = data2011[[\"q1_{}\".format(i) for i in range(1,11)]].fillna(2.5)\n",
    "X"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Normer les données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "scaler = StandardScaler()\n",
    "Xr = scaler.fit_transform(X)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Faire la projection"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.decomposition import PCA\n",
    "pca = PCA(n_components=10)\n",
    "Xp = pd.DataFrame(pca.fit_transform(Xr))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Regarder le résultat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.34376472, 0.11867907, 0.09826595, 0.08365512, 0.07389841,\n",
       "       0.06751282, 0.06464544, 0.059311  , 0.0479813 , 0.04228617])"
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pca.explained_variance_ratio_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 215,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Variance expliquée')"
      ]
     },
     "execution_count": 215,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "import numpy as np\n",
    "ind = np.arange(1, 11)\n",
    "ind\n",
    "(fig, ax) = plt.subplots(figsize=(8, 6))\n",
    "ax.plot(ind, pca.explained_variance_ratio_,\"o-\")\n",
    "ax.set_title('Éboulement')\n",
    "ax.set_xticks(ind)\n",
    "ax.set_xticklabels(ind)\n",
    "ax.set_xlabel('Composante')\n",
    "ax.set_ylabel('Variance expliquée')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Forme d'un vecteur"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1391cdc90>"
      ]
     },
     "execution_count": 178,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "pd.Series(pca.components_[0]).plot(kind=\"bar\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x135b1cb90>"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "Xp.plot(kind=\"scatter\",x=0,y=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(-1.0994713695023444,\n",
       " 1.0999748271191592,\n",
       " -1.0998615404412626,\n",
       " 1.0998615404412626)"
      ]
     },
     "execution_count": 172,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "(fig, ax) = plt.subplots(figsize=(8, 8))\n",
    "for i in range(0, pca.components_.shape[1]):\n",
    "    ax.arrow(0,0, \n",
    "             pca.components_[0, i],pca.components_[1, i],\n",
    "             head_width=0.02, head_length=0.05)\n",
    "\n",
    "    plt.text(pca.components_[0, i] + 0.07,\n",
    "             pca.components_[1, i],\n",
    "             X.columns.values[i])\n",
    "\n",
    "# Ajouter le cercle\n",
    "an = np.linspace(0, 2 * np.pi, 100)\n",
    "plt.plot(np.cos(an), np.sin(an))\n",
    "plt.axis('equal')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Classification (créer des clusters)\n",
    "\n",
    "Explorer le lien intérêt / définition\n",
    "\n",
    "- {'m_q29_1': 'Mot \"Science\" : Opinions négatives',\n",
    " 'm_q29_2': 'Mot \"Science\" : L\\'avenir / le futur',\n",
    " 'm_q29_3': 'Mot \"Science\" : La recherche (chercheurs, laboratoires)',\n",
    " 'm_q29_4': 'Mot \"Science\" : Le savoir / la connaissance / l\\'intelligence / les études',\n",
    " 'm_q29_5': 'Mot \"Science\" : Les expériences, les tests, les essais',\n",
    " 'm_q29_6': 'Mot \"Science\" : Les découvertes et inventions',\n",
    " 'm_q29_7': 'Mot \"Science\" : Le progrès / les avancées / le développement',\n",
    " 'm_q29_8': 'Mot \"Science\" : Les mathématiques',\n",
    " 'm_q29_9': 'Mot \"Science\" : La physique',\n",
    " 'm_q29_10': 'Mot \"Science\" : La chimie',\n",
    " 'm_q29_11': 'Mot \"Science\" : La biologie',\n",
    " 'm_q29_12': 'Mot \"Science\" : La génétique',\n",
    " 'm_q29_13': 'Mot \"Science\" : L\\'anatomie / le corps humain',\n",
    " 'm_q29_14': 'Mot \"Science\" : Les sciences humaines / la sociologie',\n",
    " 'm_q29_15': 'Mot \"Science\" : La météorologie',\n",
    " 'm_q29_16': 'Mot \"Science\" : Les sciences de la terre',\n",
    " 'm_q29_17': 'Mot \"Science\" : L\\'astronomie',\n",
    " 'm_q29_18': 'Mot \"Science\" : La médecine',\n",
    " 'm_q29_19': 'Mot \"Science\" : La santé / la longévité',\n",
    " 'm_q29_20': 'Mot \"Science\" : La maladie / le cancer',\n",
    " 'm_q29_21': 'Mot \"Science\" : Les médicaments / les soins / les traitements',\n",
    " 'm_q29_22': 'Mot \"Science\" : L\\'environnement',\n",
    " 'm_q29_23': 'Mot \"Science\" : Autres citations nature',\n",
    " 'm_q29_24': 'Mot \"Science\" : La technologie',\n",
    " 'm_q29_25': 'Mot \"Science\" : La technique',\n",
    " 'm_q29_26': 'Mot \"Science\" : L\\'énergie',\n",
    " 'm_q29_27': 'Mot \"Science\" : Le nucléaire / l\\'atome',\n",
    " 'm_q29_28': 'Mot \"Science\" : Autres citations',\n",
    " 'm_q29_29': 'Mot \"Science\" : Ne sait pas'}\n",
    " \n",
    " \n",
    "- 'q3_1': \"Intérêts : La conquête de l'espace\",\n",
    " 'q3_2': 'Intérêts : La recherche médicale',\n",
    " 'q3_3': 'Intérêts : Les grandes réalisations techniques',\n",
    " 'q3_4': 'Intérêts : Les grandes explorations (volcans, océans)',\n",
    " 'q3_5': 'Intérêts : Les recherches sur la matière',\n",
    " 'q3_6': 'Intérêts : Les recherches sur la vie',\n",
    " 'q3_7': \"Intérêts : Les recherches sur l'environnement\",\n",
    " 'q3_8': 'Intérêts : Les recherches sur le climat',\n",
    " 'q3_9': 'Intérêts : Les nouveautés technologiques',\n",
    " 'q3_10': 'Intérêts : Les recherches sur les énergies nouvelles',\n",
    " 'q3_11': 'Intérêts : Les risques liés aux découvertes scientifiques',"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Première étape de créer un tableau numérique"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 227,
   "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>q3_1_Beaucoup</th>\n",
       "      <th>q3_1_Ne sait pas</th>\n",
       "      <th>q3_1_Pas du tout</th>\n",
       "      <th>q3_1_Un peu</th>\n",
       "      <th>q3_2_Beaucoup</th>\n",
       "      <th>q3_2_Ne sait pas</th>\n",
       "      <th>q3_2_Pas du tout</th>\n",
       "      <th>q3_2_Un peu</th>\n",
       "      <th>q3_3_Beaucoup</th>\n",
       "      <th>q3_3_Ne sait pas</th>\n",
       "      <th>...</th>\n",
       "      <th>q3_9_Pas du tout</th>\n",
       "      <th>q3_9_Un peu</th>\n",
       "      <th>q3_10_Beaucoup</th>\n",
       "      <th>q3_10_Ne sait pas</th>\n",
       "      <th>q3_10_Pas du tout</th>\n",
       "      <th>q3_10_Un peu</th>\n",
       "      <th>q3_11_Beaucoup</th>\n",
       "      <th>q3_11_Ne sait pas</th>\n",
       "      <th>q3_11_Pas du tout</th>\n",
       "      <th>q3_11_Un peu</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1022</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1023</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1024</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1025</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1026</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1027 rows × 44 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      q3_1_Beaucoup  q3_1_Ne sait pas  q3_1_Pas du tout  q3_1_Un peu  \\\n",
       "0                 1                 0                 0            0   \n",
       "1                 0                 0                 0            1   \n",
       "2                 0                 0                 0            1   \n",
       "3                 0                 0                 1            0   \n",
       "4                 0                 0                 0            1   \n",
       "...             ...               ...               ...          ...   \n",
       "1022              1                 0                 0            0   \n",
       "1023              0                 0                 1            0   \n",
       "1024              1                 0                 0            0   \n",
       "1025              0                 0                 0            1   \n",
       "1026              0                 0                 1            0   \n",
       "\n",
       "      q3_2_Beaucoup  q3_2_Ne sait pas  q3_2_Pas du tout  q3_2_Un peu  \\\n",
       "0                 0                 0                 0            1   \n",
       "1                 0                 0                 0            1   \n",
       "2                 1                 0                 0            0   \n",
       "3                 1                 0                 0            0   \n",
       "4                 0                 0                 0            1   \n",
       "...             ...               ...               ...          ...   \n",
       "1022              0                 0                 0            1   \n",
       "1023              1                 0                 0            0   \n",
       "1024              1                 0                 0            0   \n",
       "1025              1                 0                 0            0   \n",
       "1026              1                 0                 0            0   \n",
       "\n",
       "      q3_3_Beaucoup  q3_3_Ne sait pas  ...  q3_9_Pas du tout  q3_9_Un peu  \\\n",
       "0                 0                 0  ...                 0            1   \n",
       "1                 0                 0  ...                 0            1   \n",
       "2                 1                 0  ...                 0            0   \n",
       "3                 1                 0  ...                 0            0   \n",
       "4                 1                 0  ...                 0            0   \n",
       "...             ...               ...  ...               ...          ...   \n",
       "1022              0                 0  ...                 0            1   \n",
       "1023              0                 0  ...                 0            0   \n",
       "1024              1                 0  ...                 0            0   \n",
       "1025              1                 0  ...                 0            0   \n",
       "1026              1                 0  ...                 0            0   \n",
       "\n",
       "      q3_10_Beaucoup  q3_10_Ne sait pas  q3_10_Pas du tout  q3_10_Un peu  \\\n",
       "0                  0                  0                  0             1   \n",
       "1                  0                  0                  1             0   \n",
       "2                  1                  0                  0             0   \n",
       "3                  1                  0                  0             0   \n",
       "4                  0                  0                  0             1   \n",
       "...              ...                ...                ...           ...   \n",
       "1022               0                  0                  0             1   \n",
       "1023               0                  0                  0             1   \n",
       "1024               1                  0                  0             0   \n",
       "1025               1                  0                  0             0   \n",
       "1026               1                  0                  0             0   \n",
       "\n",
       "      q3_11_Beaucoup  q3_11_Ne sait pas  q3_11_Pas du tout  q3_11_Un peu  \n",
       "0                  0                  0                  0             1  \n",
       "1                  0                  0                  1             0  \n",
       "2                  1                  0                  0             0  \n",
       "3                  0                  0                  0             1  \n",
       "4                  0                  0                  0             1  \n",
       "...              ...                ...                ...           ...  \n",
       "1022               0                  0                  0             1  \n",
       "1023               0                  0                  0             1  \n",
       "1024               1                  0                  0             0  \n",
       "1025               1                  0                  0             0  \n",
       "1026               1                  0                  0             0  \n",
       "\n",
       "[1027 rows x 44 columns]"
      ]
     },
     "execution_count": 227,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "variables = [\"q3_{}\".format(i) for i in range(1,12)]\n",
    "X = data2011[variables]\n",
    "X = X.replace(modalites(\"q3_1\",meta2011)).fillna(\"Ne sait pas\")\n",
    "Xd = pd.get_dummies(X)\n",
    "Xd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Classification hiérarchique ascendante"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 229,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2    314\n",
       "0    246\n",
       "1    214\n",
       "4    171\n",
       "3     82\n",
       "Name: cluster, dtype: int64"
      ]
     },
     "execution_count": 229,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.cluster import AgglomerativeClustering\n",
    "import numpy as np\n",
    "\n",
    "clustering = AgglomerativeClustering(n_clusters=5).fit(Xd)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ajout des clusters à nos données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data2011[\"cluster\"] = clustering.labels_\n",
    "data2011[\"cluster\"] = data2011[\"cluster\"].apply(str)\n",
    "data2011[\"cluster\"].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Croisements pour regarder la composition de ces clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 232,
   "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>cluster</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "      <th>All</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_1</th>\n",
       "      <th>1.0</th>\n",
       "      <td>42.5 (17.3%)</td>\n",
       "      <td>15.0 (7.0%)</td>\n",
       "      <td>111.7 (35.8%)</td>\n",
       "      <td>1.7 (2.1%)</td>\n",
       "      <td>34.7 (20.0%)</td>\n",
       "      <td>205.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>81.4 (33.2%)</td>\n",
       "      <td>90.6 (42.5%)</td>\n",
       "      <td>147.1 (47.1%)</td>\n",
       "      <td>11.2 (13.9%)</td>\n",
       "      <td>96.6 (55.8%)</td>\n",
       "      <td>426.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>121.2 (49.4%)</td>\n",
       "      <td>107.7 (50.5%)</td>\n",
       "      <td>53.5 (17.1%)</td>\n",
       "      <td>67.9 (84.0%)</td>\n",
       "      <td>41.8 (24.1%)</td>\n",
       "      <td>392.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>245.1</td>\n",
       "      <td>213.3</td>\n",
       "      <td>312.3</td>\n",
       "      <td>80.8</td>\n",
       "      <td>173.2</td>\n",
       "      <td>1024.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_2</th>\n",
       "      <th>1.0</th>\n",
       "      <td>196.2 (79.6%)</td>\n",
       "      <td>102.8 (48.2%)</td>\n",
       "      <td>283.7 (90.8%)</td>\n",
       "      <td>34.1 (42.2%)</td>\n",
       "      <td>118.8 (68.6%)</td>\n",
       "      <td>735.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>41.7 (16.9%)</td>\n",
       "      <td>104.7 (49.1%)</td>\n",
       "      <td>27.0 (8.6%)</td>\n",
       "      <td>38.4 (47.5%)</td>\n",
       "      <td>53.4 (30.8%)</td>\n",
       "      <td>265.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>8.5 (3.4%)</td>\n",
       "      <td>5.8 (2.7%)</td>\n",
       "      <td>1.6 (0.5%)</td>\n",
       "      <td>8.3 (10.3%)</td>\n",
       "      <td>1.0 (0.6%)</td>\n",
       "      <td>25.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>246.4</td>\n",
       "      <td>213.3</td>\n",
       "      <td>312.3</td>\n",
       "      <td>80.8</td>\n",
       "      <td>173.2</td>\n",
       "      <td>1026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_3</th>\n",
       "      <th>1.0</th>\n",
       "      <td>45.4 (18.8%)</td>\n",
       "      <td>11.3 (5.4%)</td>\n",
       "      <td>269.7 (86.3%)</td>\n",
       "      <td>3.0 (3.7%)</td>\n",
       "      <td>93.1 (54.2%)</td>\n",
       "      <td>422.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>172.4 (71.4%)</td>\n",
       "      <td>167.3 (79.3%)</td>\n",
       "      <td>38.3 (12.3%)</td>\n",
       "      <td>32.9 (41.1%)</td>\n",
       "      <td>69.5 (40.4%)</td>\n",
       "      <td>480.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>23.8 (9.9%)</td>\n",
       "      <td>32.5 (15.4%)</td>\n",
       "      <td>4.4 (1.4%)</td>\n",
       "      <td>44.2 (55.2%)</td>\n",
       "      <td>9.3 (5.4%)</td>\n",
       "      <td>114.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>241.6</td>\n",
       "      <td>211.1</td>\n",
       "      <td>312.3</td>\n",
       "      <td>80</td>\n",
       "      <td>171.8</td>\n",
       "      <td>1016.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_4</th>\n",
       "      <th>1.0</th>\n",
       "      <td>96.1 (39.0%)</td>\n",
       "      <td>19.9 (9.4%)</td>\n",
       "      <td>220.0 (70.6%)</td>\n",
       "      <td>1.2 (1.5%)</td>\n",
       "      <td>64.5 (37.2%)</td>\n",
       "      <td>401.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>105.0 (42.6%)</td>\n",
       "      <td>142.3 (67.3%)</td>\n",
       "      <td>82.7 (26.5%)</td>\n",
       "      <td>23.4 (29.2%)</td>\n",
       "      <td>82.5 (47.6%)</td>\n",
       "      <td>435.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>45.4 (18.4%)</td>\n",
       "      <td>49.2 (23.3%)</td>\n",
       "      <td>9.0 (2.9%)</td>\n",
       "      <td>55.4 (69.2%)</td>\n",
       "      <td>26.2 (15.1%)</td>\n",
       "      <td>185.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>246.4</td>\n",
       "      <td>211.4</td>\n",
       "      <td>311.7</td>\n",
       "      <td>80</td>\n",
       "      <td>173.2</td>\n",
       "      <td>1022.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_5</th>\n",
       "      <th>1.0</th>\n",
       "      <td>53.9 (23.6%)</td>\n",
       "      <td>5.8 (2.8%)</td>\n",
       "      <td>178.5 (57.8%)</td>\n",
       "      <td>0.0 (0.0%)</td>\n",
       "      <td>59.0 (34.8%)</td>\n",
       "      <td>297.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>110.6 (48.5%)</td>\n",
       "      <td>106.5 (51.5%)</td>\n",
       "      <td>106.4 (34.5%)</td>\n",
       "      <td>12.1 (15.0%)</td>\n",
       "      <td>90.8 (53.6%)</td>\n",
       "      <td>426.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>63.7 (27.9%)</td>\n",
       "      <td>94.5 (45.7%)</td>\n",
       "      <td>23.7 (7.7%)</td>\n",
       "      <td>68.7 (85.0%)</td>\n",
       "      <td>19.6 (11.6%)</td>\n",
       "      <td>270.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>228.2</td>\n",
       "      <td>206.8</td>\n",
       "      <td>308.6</td>\n",
       "      <td>80.8</td>\n",
       "      <td>169.4</td>\n",
       "      <td>993.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_6</th>\n",
       "      <th>1.0</th>\n",
       "      <td>177.1 (73.8%)</td>\n",
       "      <td>42.4 (20.0%)</td>\n",
       "      <td>280.9 (90.6%)</td>\n",
       "      <td>8.8 (10.9%)</td>\n",
       "      <td>100.3 (57.9%)</td>\n",
       "      <td>609.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>54.2 (22.6%)</td>\n",
       "      <td>147.6 (69.5%)</td>\n",
       "      <td>29.3 (9.4%)</td>\n",
       "      <td>43.7 (54.1%)</td>\n",
       "      <td>69.5 (40.2%)</td>\n",
       "      <td>344.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>8.8 (3.7%)</td>\n",
       "      <td>22.3 (10.5%)</td>\n",
       "      <td>0.0 (0.0%)</td>\n",
       "      <td>28.3 (35.0%)</td>\n",
       "      <td>3.3 (1.9%)</td>\n",
       "      <td>62.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>240.1</td>\n",
       "      <td>212.3</td>\n",
       "      <td>310.2</td>\n",
       "      <td>80.8</td>\n",
       "      <td>173.2</td>\n",
       "      <td>1016.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_7</th>\n",
       "      <th>1.0</th>\n",
       "      <td>217.0 (89.2%)</td>\n",
       "      <td>15.5 (7.3%)</td>\n",
       "      <td>296.2 (94.8%)</td>\n",
       "      <td>2.4 (3.0%)</td>\n",
       "      <td>63.9 (36.9%)</td>\n",
       "      <td>595</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>12.6 (5.2%)</td>\n",
       "      <td>195.0 (91.4%)</td>\n",
       "      <td>16.2 (5.2%)</td>\n",
       "      <td>42.8 (53.0%)</td>\n",
       "      <td>99.9 (57.7%)</td>\n",
       "      <td>366.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>13.7 (5.6%)</td>\n",
       "      <td>2.9 (1.4%)</td>\n",
       "      <td>0.0 (0.0%)</td>\n",
       "      <td>35.6 (44.1%)</td>\n",
       "      <td>9.4 (5.4%)</td>\n",
       "      <td>61.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>243.2</td>\n",
       "      <td>213.3</td>\n",
       "      <td>312.3</td>\n",
       "      <td>80.8</td>\n",
       "      <td>173.2</td>\n",
       "      <td>1022.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_8</th>\n",
       "      <th>1.0</th>\n",
       "      <td>200.6 (82.4%)</td>\n",
       "      <td>14.6 (6.9%)</td>\n",
       "      <td>289.6 (92.7%)</td>\n",
       "      <td>5.6 (6.9%)</td>\n",
       "      <td>27.3 (15.9%)</td>\n",
       "      <td>537.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>29.4 (12.1%)</td>\n",
       "      <td>182.8 (86.1%)</td>\n",
       "      <td>22.8 (7.3%)</td>\n",
       "      <td>26.3 (32.5%)</td>\n",
       "      <td>128.7 (74.7%)</td>\n",
       "      <td>390</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>13.3 (5.5%)</td>\n",
       "      <td>14.8 (7.0%)</td>\n",
       "      <td>0.0 (0.0%)</td>\n",
       "      <td>48.9 (60.5%)</td>\n",
       "      <td>16.2 (9.4%)</td>\n",
       "      <td>93.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>243.2</td>\n",
       "      <td>212.2</td>\n",
       "      <td>312.3</td>\n",
       "      <td>80.8</td>\n",
       "      <td>172.3</td>\n",
       "      <td>1020.8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_9</th>\n",
       "      <th>1.0</th>\n",
       "      <td>39.0 (15.8%)</td>\n",
       "      <td>17.7 (8.3%)</td>\n",
       "      <td>265.7 (85.1%)</td>\n",
       "      <td>8.3 (10.3%)</td>\n",
       "      <td>131.1 (75.7%)</td>\n",
       "      <td>461.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>179.4 (72.8%)</td>\n",
       "      <td>155.5 (73.2%)</td>\n",
       "      <td>37.6 (12.0%)</td>\n",
       "      <td>20.8 (25.7%)</td>\n",
       "      <td>35.7 (20.6%)</td>\n",
       "      <td>429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>28.1 (11.4%)</td>\n",
       "      <td>39.2 (18.5%)</td>\n",
       "      <td>9.0 (2.9%)</td>\n",
       "      <td>51.7 (64.0%)</td>\n",
       "      <td>6.4 (3.7%)</td>\n",
       "      <td>134.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>246.4</td>\n",
       "      <td>212.5</td>\n",
       "      <td>312.3</td>\n",
       "      <td>80.8</td>\n",
       "      <td>173.2</td>\n",
       "      <td>1025.2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_10</th>\n",
       "      <th>1.0</th>\n",
       "      <td>128.0 (53.2%)</td>\n",
       "      <td>23.4 (11.0%)</td>\n",
       "      <td>304.6 (97.5%)</td>\n",
       "      <td>1.0 (1.2%)</td>\n",
       "      <td>112.3 (64.8%)</td>\n",
       "      <td>569.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>101.0 (41.9%)</td>\n",
       "      <td>176.1 (82.6%)</td>\n",
       "      <td>5.9 (1.9%)</td>\n",
       "      <td>23.8 (29.5%)</td>\n",
       "      <td>57.3 (33.1%)</td>\n",
       "      <td>364.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>11.8 (4.9%)</td>\n",
       "      <td>13.8 (6.5%)</td>\n",
       "      <td>1.8 (0.6%)</td>\n",
       "      <td>56.0 (69.3%)</td>\n",
       "      <td>3.6 (2.1%)</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>240.9</td>\n",
       "      <td>213.3</td>\n",
       "      <td>312.3</td>\n",
       "      <td>80.8</td>\n",
       "      <td>173.2</td>\n",
       "      <td>1020.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"4\" valign=\"top\">q3_11</th>\n",
       "      <th>1.0</th>\n",
       "      <td>132.0 (54.4%)</td>\n",
       "      <td>30.3 (14.3%)</td>\n",
       "      <td>257.8 (82.8%)</td>\n",
       "      <td>2.6 (3.2%)</td>\n",
       "      <td>82.2 (47.5%)</td>\n",
       "      <td>504.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2.0</th>\n",
       "      <td>86.3 (35.6%)</td>\n",
       "      <td>164.9 (77.7%)</td>\n",
       "      <td>48.2 (15.5%)</td>\n",
       "      <td>10.7 (13.2%)</td>\n",
       "      <td>81.4 (47.0%)</td>\n",
       "      <td>391.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3.0</th>\n",
       "      <td>24.4 (10.1%)</td>\n",
       "      <td>17.1 (8.1%)</td>\n",
       "      <td>5.4 (1.7%)</td>\n",
       "      <td>67.5 (83.5%)</td>\n",
       "      <td>9.6 (5.5%)</td>\n",
       "      <td>124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>242.7</td>\n",
       "      <td>212.3</td>\n",
       "      <td>311.5</td>\n",
       "      <td>80.8</td>\n",
       "      <td>173.2</td>\n",
       "      <td>1020.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"3\" valign=\"top\">sexe</th>\n",
       "      <th>Femme</th>\n",
       "      <td>150.9 (61.0%)</td>\n",
       "      <td>127.0 (59.5%)</td>\n",
       "      <td>136.6 (43.7%)</td>\n",
       "      <td>51.5 (63.7%)</td>\n",
       "      <td>72.1 (41.6%)</td>\n",
       "      <td>538.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Homme</th>\n",
       "      <td>96.5 (39.0%)</td>\n",
       "      <td>86.4 (40.5%)</td>\n",
       "      <td>175.7 (56.3%)</td>\n",
       "      <td>29.3 (36.3%)</td>\n",
       "      <td>101.1 (58.4%)</td>\n",
       "      <td>488.9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>All</th>\n",
       "      <td>247.4</td>\n",
       "      <td>213.3</td>\n",
       "      <td>312.3</td>\n",
       "      <td>80.8</td>\n",
       "      <td>173.2</td>\n",
       "      <td>1027</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "cluster                  0              1              2             3  \\\n",
       "q3_1  1.0     42.5 (17.3%)    15.0 (7.0%)  111.7 (35.8%)    1.7 (2.1%)   \n",
       "      2.0     81.4 (33.2%)   90.6 (42.5%)  147.1 (47.1%)  11.2 (13.9%)   \n",
       "      3.0    121.2 (49.4%)  107.7 (50.5%)   53.5 (17.1%)  67.9 (84.0%)   \n",
       "      All            245.1          213.3          312.3          80.8   \n",
       "q3_2  1.0    196.2 (79.6%)  102.8 (48.2%)  283.7 (90.8%)  34.1 (42.2%)   \n",
       "      2.0     41.7 (16.9%)  104.7 (49.1%)    27.0 (8.6%)  38.4 (47.5%)   \n",
       "      3.0       8.5 (3.4%)     5.8 (2.7%)     1.6 (0.5%)   8.3 (10.3%)   \n",
       "      All            246.4          213.3          312.3          80.8   \n",
       "q3_3  1.0     45.4 (18.8%)    11.3 (5.4%)  269.7 (86.3%)    3.0 (3.7%)   \n",
       "      2.0    172.4 (71.4%)  167.3 (79.3%)   38.3 (12.3%)  32.9 (41.1%)   \n",
       "      3.0      23.8 (9.9%)   32.5 (15.4%)     4.4 (1.4%)  44.2 (55.2%)   \n",
       "      All            241.6          211.1          312.3            80   \n",
       "q3_4  1.0     96.1 (39.0%)    19.9 (9.4%)  220.0 (70.6%)    1.2 (1.5%)   \n",
       "      2.0    105.0 (42.6%)  142.3 (67.3%)   82.7 (26.5%)  23.4 (29.2%)   \n",
       "      3.0     45.4 (18.4%)   49.2 (23.3%)     9.0 (2.9%)  55.4 (69.2%)   \n",
       "      All            246.4          211.4          311.7            80   \n",
       "q3_5  1.0     53.9 (23.6%)     5.8 (2.8%)  178.5 (57.8%)    0.0 (0.0%)   \n",
       "      2.0    110.6 (48.5%)  106.5 (51.5%)  106.4 (34.5%)  12.1 (15.0%)   \n",
       "      3.0     63.7 (27.9%)   94.5 (45.7%)    23.7 (7.7%)  68.7 (85.0%)   \n",
       "      All            228.2          206.8          308.6          80.8   \n",
       "q3_6  1.0    177.1 (73.8%)   42.4 (20.0%)  280.9 (90.6%)   8.8 (10.9%)   \n",
       "      2.0     54.2 (22.6%)  147.6 (69.5%)    29.3 (9.4%)  43.7 (54.1%)   \n",
       "      3.0       8.8 (3.7%)   22.3 (10.5%)     0.0 (0.0%)  28.3 (35.0%)   \n",
       "      All            240.1          212.3          310.2          80.8   \n",
       "q3_7  1.0    217.0 (89.2%)    15.5 (7.3%)  296.2 (94.8%)    2.4 (3.0%)   \n",
       "      2.0      12.6 (5.2%)  195.0 (91.4%)    16.2 (5.2%)  42.8 (53.0%)   \n",
       "      3.0      13.7 (5.6%)     2.9 (1.4%)     0.0 (0.0%)  35.6 (44.1%)   \n",
       "      All            243.2          213.3          312.3          80.8   \n",
       "q3_8  1.0    200.6 (82.4%)    14.6 (6.9%)  289.6 (92.7%)    5.6 (6.9%)   \n",
       "      2.0     29.4 (12.1%)  182.8 (86.1%)    22.8 (7.3%)  26.3 (32.5%)   \n",
       "      3.0      13.3 (5.5%)    14.8 (7.0%)     0.0 (0.0%)  48.9 (60.5%)   \n",
       "      All            243.2          212.2          312.3          80.8   \n",
       "q3_9  1.0     39.0 (15.8%)    17.7 (8.3%)  265.7 (85.1%)   8.3 (10.3%)   \n",
       "      2.0    179.4 (72.8%)  155.5 (73.2%)   37.6 (12.0%)  20.8 (25.7%)   \n",
       "      3.0     28.1 (11.4%)   39.2 (18.5%)     9.0 (2.9%)  51.7 (64.0%)   \n",
       "      All            246.4          212.5          312.3          80.8   \n",
       "q3_10 1.0    128.0 (53.2%)   23.4 (11.0%)  304.6 (97.5%)    1.0 (1.2%)   \n",
       "      2.0    101.0 (41.9%)  176.1 (82.6%)     5.9 (1.9%)  23.8 (29.5%)   \n",
       "      3.0      11.8 (4.9%)    13.8 (6.5%)     1.8 (0.6%)  56.0 (69.3%)   \n",
       "      All            240.9          213.3          312.3          80.8   \n",
       "q3_11 1.0    132.0 (54.4%)   30.3 (14.3%)  257.8 (82.8%)    2.6 (3.2%)   \n",
       "      2.0     86.3 (35.6%)  164.9 (77.7%)   48.2 (15.5%)  10.7 (13.2%)   \n",
       "      3.0     24.4 (10.1%)    17.1 (8.1%)     5.4 (1.7%)  67.5 (83.5%)   \n",
       "      All            242.7          212.3          311.5          80.8   \n",
       "sexe  Femme  150.9 (61.0%)  127.0 (59.5%)  136.6 (43.7%)  51.5 (63.7%)   \n",
       "      Homme   96.5 (39.0%)   86.4 (40.5%)  175.7 (56.3%)  29.3 (36.3%)   \n",
       "      All            247.4          213.3          312.3          80.8   \n",
       "\n",
       "cluster                  4     All  \n",
       "q3_1  1.0     34.7 (20.0%)   205.6  \n",
       "      2.0     96.6 (55.8%)   426.9  \n",
       "      3.0     41.8 (24.1%)   392.2  \n",
       "      All            173.2  1024.7  \n",
       "q3_2  1.0    118.8 (68.6%)   735.6  \n",
       "      2.0     53.4 (30.8%)   265.2  \n",
       "      3.0       1.0 (0.6%)    25.2  \n",
       "      All            173.2    1026  \n",
       "q3_3  1.0     93.1 (54.2%)   422.4  \n",
       "      2.0     69.5 (40.4%)   480.4  \n",
       "      3.0       9.3 (5.4%)   114.1  \n",
       "      All            171.8  1016.8  \n",
       "q3_4  1.0     64.5 (37.2%)   401.6  \n",
       "      2.0     82.5 (47.6%)   435.8  \n",
       "      3.0     26.2 (15.1%)   185.2  \n",
       "      All            173.2  1022.6  \n",
       "q3_5  1.0     59.0 (34.8%)   297.2  \n",
       "      2.0     90.8 (53.6%)   426.3  \n",
       "      3.0     19.6 (11.6%)   270.3  \n",
       "      All            169.4   993.8  \n",
       "q3_6  1.0    100.3 (57.9%)   609.6  \n",
       "      2.0     69.5 (40.2%)   344.3  \n",
       "      3.0       3.3 (1.9%)    62.7  \n",
       "      All            173.2  1016.6  \n",
       "q3_7  1.0     63.9 (36.9%)     595  \n",
       "      2.0     99.9 (57.7%)   366.4  \n",
       "      3.0       9.4 (5.4%)    61.5  \n",
       "      All            173.2  1022.8  \n",
       "q3_8  1.0     27.3 (15.9%)   537.7  \n",
       "      2.0    128.7 (74.7%)     390  \n",
       "      3.0      16.2 (9.4%)    93.1  \n",
       "      All            172.3  1020.8  \n",
       "q3_9  1.0    131.1 (75.7%)   461.9  \n",
       "      2.0     35.7 (20.6%)     429  \n",
       "      3.0       6.4 (3.7%)   134.3  \n",
       "      All            173.2  1025.2  \n",
       "q3_10 1.0    112.3 (64.8%)   569.3  \n",
       "      2.0     57.3 (33.1%)   364.1  \n",
       "      3.0       3.6 (2.1%)      87  \n",
       "      All            173.2  1020.5  \n",
       "q3_11 1.0     82.2 (47.5%)   504.9  \n",
       "      2.0     81.4 (47.0%)   391.5  \n",
       "      3.0       9.6 (5.5%)     124  \n",
       "      All            173.2  1020.4  \n",
       "sexe  Femme   72.1 (41.6%)   538.1  \n",
       "      Homme  101.1 (58.4%)   488.9  \n",
       "      All            173.2    1027  "
      ]
     },
     "execution_count": 232,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "t = {}\n",
    "for i in variables + [\"sexe\"]:\n",
    "    t[i] = pyshs.tableau_croise(data2011, \"cluster\",i, \"poids1\").T\n",
    "pd.concat(t)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Essayer d'analyser les sens donné au terme science"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Créer le tableau de données"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 287,
   "metadata": {},
   "outputs": [],
   "source": [
    "science = {'m_q29_1': 'Mot \"Science\" : Opinions négatives',\n",
    " 'm_q29_2': 'Mot \"Science\" : L\\'avenir / le futur',\n",
    " 'm_q29_3': 'Mot \"Science\" : La recherche (chercheurs, laboratoires)',\n",
    " 'm_q29_4': 'Mot \"Science\" : Le savoir / la connaissance / l\\'intelligence / les études',\n",
    " 'm_q29_5': 'Mot \"Science\" : Les expériences, les tests, les essais',\n",
    " 'm_q29_6': 'Mot \"Science\" : Les découvertes et inventions',\n",
    " 'm_q29_7': 'Mot \"Science\" : Le progrès / les avancées / le développement',\n",
    " 'm_q29_8': 'Mot \"Science\" : Les mathématiques',\n",
    " 'm_q29_9': 'Mot \"Science\" : La physique',\n",
    " 'm_q29_10': 'Mot \"Science\" : La chimie',\n",
    " 'm_q29_11': 'Mot \"Science\" : La biologie',\n",
    " 'm_q29_12': 'Mot \"Science\" : La génétique',\n",
    " 'm_q29_13': 'Mot \"Science\" : L\\'anatomie / le corps humain',\n",
    " 'm_q29_14': 'Mot \"Science\" : Les sciences humaines / la sociologie',\n",
    " 'm_q29_15': 'Mot \"Science\" : La météorologie',\n",
    " 'm_q29_16': 'Mot \"Science\" : Les sciences de la terre',\n",
    " 'm_q29_17': 'Mot \"Science\" : L\\'astronomie',\n",
    " 'm_q29_18': 'Mot \"Science\" : La médecine',\n",
    " 'm_q29_19': 'Mot \"Science\" : La santé / la longévité',\n",
    " 'm_q29_20': 'Mot \"Science\" : La maladie / le cancer',\n",
    " 'm_q29_21': 'Mot \"Science\" : Les médicaments / les soins / les traitements',\n",
    " 'm_q29_22': 'Mot \"Science\" : L\\'environnement',\n",
    " 'm_q29_23': 'Mot \"Science\" : Autres citations nature',\n",
    " 'm_q29_24': 'Mot \"Science\" : La technologie',\n",
    " 'm_q29_25': 'Mot \"Science\" : La technique',\n",
    " 'm_q29_26': 'Mot \"Science\" : L\\'énergie',\n",
    " 'm_q29_27': 'Mot \"Science\" : Le nucléaire / l\\'atome',\n",
    " 'm_q29_28': 'Mot \"Science\" : Autres citations',\n",
    " 'm_q29_29': 'Mot \"Science\" : Ne sait pas'}\n",
    "science = {i:science[i].replace('Mot \"Science\" : ','') for i in science}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{0.0: 'Non cité', 1.0: 'Cité'}"
      ]
     },
     "execution_count": 275,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "modalites(\"m_q29_1\",meta2011)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 295,
   "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>Opinions négatives</th>\n",
       "      <th>L'avenir / le futur</th>\n",
       "      <th>La recherche (chercheurs, laboratoires)</th>\n",
       "      <th>Le savoir / la connaissance / l'intelligence / les études</th>\n",
       "      <th>Les expériences, les tests, les essais</th>\n",
       "      <th>Les découvertes et inventions</th>\n",
       "      <th>Le progrès / les avancées / le développement</th>\n",
       "      <th>Les mathématiques</th>\n",
       "      <th>La physique</th>\n",
       "      <th>La chimie</th>\n",
       "      <th>...</th>\n",
       "      <th>La maladie / le cancer</th>\n",
       "      <th>Les médicaments / les soins / les traitements</th>\n",
       "      <th>L'environnement</th>\n",
       "      <th>Autres citations nature</th>\n",
       "      <th>La technologie</th>\n",
       "      <th>La technique</th>\n",
       "      <th>L'énergie</th>\n",
       "      <th>Le nucléaire / l'atome</th>\n",
       "      <th>Autres citations</th>\n",
       "      <th>Ne sait pas</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1022</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1023</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1024</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1025</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1026</th>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>...</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Oui</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "      <td>Non</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1027 rows × 29 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     Opinions négatives L'avenir / le futur  \\\n",
       "0                   Non                 Non   \n",
       "1                   Non                 Non   \n",
       "2                   Non                 Non   \n",
       "3                   Non                 Non   \n",
       "4                   Non                 Non   \n",
       "...                 ...                 ...   \n",
       "1022                Non                 Non   \n",
       "1023                Non                 Non   \n",
       "1024                Non                 Non   \n",
       "1025                Non                 Non   \n",
       "1026                Non                 Non   \n",
       "\n",
       "     La recherche (chercheurs, laboratoires)  \\\n",
       "0                                        Non   \n",
       "1                                        Non   \n",
       "2                                        Non   \n",
       "3                                        Non   \n",
       "4                                        Oui   \n",
       "...                                      ...   \n",
       "1022                                     Oui   \n",
       "1023                                     Oui   \n",
       "1024                                     Oui   \n",
       "1025                                     Non   \n",
       "1026                                     Non   \n",
       "\n",
       "     Le savoir / la connaissance / l'intelligence / les études  \\\n",
       "0                                                   Non          \n",
       "1                                                   Non          \n",
       "2                                                   Non          \n",
       "3                                                   Non          \n",
       "4                                                   Non          \n",
       "...                                                 ...          \n",
       "1022                                                Non          \n",
       "1023                                                Non          \n",
       "1024                                                Non          \n",
       "1025                                                Non          \n",
       "1026                                                Non          \n",
       "\n",
       "     Les expériences, les tests, les essais Les découvertes et inventions  \\\n",
       "0                                       Non                           Non   \n",
       "1                                       Non                           Non   \n",
       "2                                       Non                           Non   \n",
       "3                                       Non                           Non   \n",
       "4                                       Non                           Non   \n",
       "...                                     ...                           ...   \n",
       "1022                                    Non                           Non   \n",
       "1023                                    Non                           Non   \n",
       "1024                                    Non                           Non   \n",
       "1025                                    Oui                           Non   \n",
       "1026                                    Non                           Non   \n",
       "\n",
       "     Le progrès / les avancées / le développement Les mathématiques  \\\n",
       "0                                             Non               Non   \n",
       "1                                             Non               Non   \n",
       "2                                             Non               Non   \n",
       "3                                             Oui               Non   \n",
       "4                                             Non               Non   \n",
       "...                                           ...               ...   \n",
       "1022                                          Non               Non   \n",
       "1023                                          Non               Non   \n",
       "1024                                          Non               Non   \n",
       "1025                                          Non               Non   \n",
       "1026                                          Non               Non   \n",
       "\n",
       "     La physique La chimie  ... La maladie / le cancer  \\\n",
       "0            Non       Non  ...                    Non   \n",
       "1            Non       Non  ...                    Non   \n",
       "2            Non       Non  ...                    Non   \n",
       "3            Non       Non  ...                    Non   \n",
       "4            Non       Non  ...                    Non   \n",
       "...          ...       ...  ...                    ...   \n",
       "1022         Non       Non  ...                    Non   \n",
       "1023         Non       Non  ...                    Oui   \n",
       "1024         Non       Non  ...                    Oui   \n",
       "1025         Non       Non  ...                    Non   \n",
       "1026         Non       Non  ...                    Non   \n",
       "\n",
       "     Les médicaments / les soins / les traitements L'environnement  \\\n",
       "0                                              Non             Non   \n",
       "1                                              Non             Non   \n",
       "2                                              Non             Non   \n",
       "3                                              Non             Non   \n",
       "4                                              Non             Non   \n",
       "...                                            ...             ...   \n",
       "1022                                           Non             Non   \n",
       "1023                                           Non             Non   \n",
       "1024                                           Non             Non   \n",
       "1025                                           Non             Non   \n",
       "1026                                           Oui             Non   \n",
       "\n",
       "     Autres citations nature La technologie La technique L'énergie  \\\n",
       "0                        Non            Non          Non       Non   \n",
       "1                        Non            Non          Non       Non   \n",
       "2                        Non            Non          Non       Non   \n",
       "3                        Non            Oui          Non       Non   \n",
       "4                        Non            Non          Non       Non   \n",
       "...                      ...            ...          ...       ...   \n",
       "1022                     Non            Non          Non       Non   \n",
       "1023                     Non            Non          Non       Non   \n",
       "1024                     Non            Oui          Non       Non   \n",
       "1025                     Non            Non          Non       Non   \n",
       "1026                     Non            Oui          Non       Non   \n",
       "\n",
       "     Le nucléaire / l'atome Autres citations Ne sait pas  \n",
       "0                       Non              Oui         Non  \n",
       "1                       Non              Non         Non  \n",
       "2                       Non              Non         Non  \n",
       "3                       Non              Oui         Non  \n",
       "4                       Non              Non         Non  \n",
       "...                     ...              ...         ...  \n",
       "1022                    Non              Non         Non  \n",
       "1023                    Non              Non         Non  \n",
       "1024                    Non              Non         Non  \n",
       "1025                    Non              Non         Non  \n",
       "1026                    Non              Non         Non  \n",
       "\n",
       "[1027 rows x 29 columns]"
      ]
     },
     "execution_count": 295,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X = data2011[[i for i in science]]\n",
    "X.columns = list(science.values())\n",
    "X = X.replace({1:\"Oui\",0:\"Non\"})\n",
    "X"
   ]
  }
 ],
 "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": "3.7.7"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}