ENEM_1999-2008.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# ENEM: Analyzing data from 1999 to 2008\n",
    "_by Lucas Bracher_\n",
    "\n",
    "ENEM, Brazilian National High School Exam, was in its first 11 years an evaluation tool for Brazilian high school. At the end of this period, it became a compulsory examination for anyone who wanted to enter federal colleges and universities, that is, it became an examination similar to the Italian *Esame di Maturità* and the French *Baccalauréat*. Fortunately, Brazilian Ministry of Education made available the scores and socioeconomic data of each candidate, so that we can make very interesting data crossings of these first 10 years, as well as discover relationships between candidates' grades with family income, presence of tap water and electricity in house, as well as the relation between the scores and the studies in public or private school and its relation with concomitant work during high school. In this article we'll look at these files with the [Pandas](https://pandas.pydata.org/), [NumPy](http://www.numpy.org/) and [Matplotlib](https://matplotlib.org /) libraries.\n",
    "\n",
    "## Getting the data\n",
    "\n",
    "The data can be downloaded from the [INEP website](http://inep.gov.br/microdados), institute responsible for the elaboration, application and organization of the results. All data available up to March 14, 2019 goes from the first edition (1998) until 2017. Unfortunately, because few students answered to the socio-economic evaluation of the 1998 edition, we decided not to use the data from this edition. Therefore, we used the data from 1999 to 2008. From 2009 the data became available in a slightly more complex format and we thought it would be better to analyze them in a second moment.\n",
    "\n",
    "The data for each of these first years are made available in `.sas` and` .sps` format, which are data description files, related to another text file with fixed width fields.\n",
    "\n",
    "## Analysis of the content of each year and preparation of the data\n",
    "\n",
    "In general, each set of data for each year is made available in a `.zip` file containing the following directories:\n",
    "\n",
    "* `DATA`: contains a` .txt` file containing the notes and socio-economic data of each student. This file is in fixed width field format.\n",
    "* `INPUT_SAS_SPSS`: contains the metadata files for reading the` .txt` file contained in `DATA`.\n",
    "* `README `: contains a` .pdf` extension file containing the data dictionary for reading the `.txt` file, as well as the contents of the tests.\n",
    "\n",
    "For the use of the data of each year, the `.txt` and` .sas` files of each year were placed in the same folder. Each `.sas` file was edited so that the` .txt` file could be read while it was in the same folder, because originally the required path is `C: \\`.\n",
    "\n",
    "## First problems\n",
    "\n",
    "The first problem we faced was that the `.sas` files could not be read with the package sas made available in Python:\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "```python\n",
    "import pandas as pd\n",
    "enem1998 = pd.read_sas(\"INPUT_SAS_ENEM_1998.sas\")\n",
    "\n",
    "---------------------------------------------------------------------------\n",
    "ValueError                                Traceback (most recent call last)\n",
    "<ipython-input-1-c40e8ad19bdf> in <module>\n",
    "      1 import pandas as pd\n",
    "      2 \n",
    "----> 3 enem1998 = pd.read_sas(\"INPUT_SAS_ENEM_1998.sas\")\n",
    "\n",
    "~/.local/lib/python3.6/site-packages/pandas/io/sas/sasreader.py in read_sas(filepath_or_buffer, format, index, encoding, chunksize, iterator)\n",
    "     46             format = \"sas7bdat\"\n",
    "     47         else:\n",
    "---> 48             raise ValueError(\"unable to infer format of SAS file\")\n",
    "     49 \n",
    "     50     if format.lower() == 'xport':\n",
    "\n",
    "ValueError: unable to infer format of SAS file\n",
    "```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The second problem is the inconsistency of fields and data over the years. There are several fields in the data of each year whose names have no clear meaning; each year such fields have different names, different positions and different contents, apart from the fact that the Pandas `.csv` file parser sometimes does not make a good guess of which would be the best type for each column, resulting a DataFrame of much larger size than necessary.\n",
    "\n",
    "## Solutions\n",
    "\n",
    "The solution found for the first problem was to write a quick \"parser\" based on .sas files. To build this parser, we opened an .sas file and we got the following in the first few lines:\n",
    "\n",
    "```\n",
    "DATE ENEM_1998;\n",
    "RETAIN MASC_INSCRITO NU_ANO DT_NASCIMENTO;\n",
    "INFILE \"C: \\ DATA_ENEM_1998.TXT\"\n",
    "LRECL = 477 MISSOVER;\n",
    "INPUT\n",
    "@1 MASC_INSCRITO $ Char8. / * Enrollment Mask * /\n",
    "@9 NU_ANO 8. / * Year of the Enem * /\n",
    "@17 DT_NASCIMENTO $ Char20. / * Date of birth of the registrant * /\n",
    "@37 TP_SEXO 8. / * Sex of member * /\n",
    "@45 CODMUNIC_INSC $ Char12. / * Municipal Code in which the enrollee lives: * /\n",
    ".\n",
    ".\n",
    ".\n",
    "```\n",
    "\n",
    "From these files a quick parser was written, and it will be presented just below.\n",
    "\n",
    "The solution to the second problem was to construct a metadata table containing the information of each field in each year in order to normalize both column names and contents. This table contains 6 fields:\n",
    "\n",
    "* `YEAR`: year of the ENEM test\n",
    "\n",
    "* `FIELD`: field of ENEM data year file\n",
    "\n",
    "* `TYPE`: type to be converted to reading `.csv` file\n",
    "\n",
    "* `RENAME_TO`: significant name for which the field will be renamed\n",
    "\n",
    "* `MANDATORY`: If this field is set to 1, only records containing this field will be loaded.\n",
    "\n",
    "* `CONVERT_TO`: data conversion dictionary for more meaningful content.\n",
    "\n",
    "Such a table was written to a `.csv` file and loaded using Pandas itself.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Downloading and processing data data (optional step)\n",
    "\n",
    "This step is optional and can last several hours of processing and can be skipped. You can go to the [Fast Data Loading](#Fast-data-loading) part, where data is loaded from a binary file in a few seconds.\n",
    "\n",
    "In order to download the data from the internet and make it into `.csv` files, just run the code below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "909b61a640f34df18d8fb5be0549f2cf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HTML(value='', placeholder='')"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import collections\n",
    "import csv, io, zipfile\n",
    "import requests\n",
    "import re\n",
    "import sys, os, shutil\n",
    "from IPython.display import display\n",
    "import ipywidgets as widgets\n",
    "\n",
    "out = widgets.HTML(value='',\n",
    "    placeholder='',\n",
    "    description='')\n",
    "display(out)\n",
    "\n",
    "urlprefix = 'http://download.inep.gov.br/microdados/'\n",
    "\n",
    "files = {\n",
    "    'micro_enem1999.zip' : \n",
    "        {'meta' : 'IMPUT_SAS_SPSS/INPUT_SAS_ENEM_1999.sas', \n",
    "         'data' : 'DADOS/DADOS_ENEM_1999.zip', \n",
    "         'rename' : 'DADOS_ENEM_1999.csv'},\n",
    "    'micro_enem2000.zip' : \n",
    "        {'meta' : 'IMPUT_SAS_SPSS/INPUT_SAS_ENEM_2000.sas',\n",
    "         'data' : 'DADOS/DADOS_ENEM_2000.zip', \n",
    "         'rename' : 'DADOS_ENEM_2000.csv'},\n",
    "    'micro_enem2001.zip' : \n",
    "        {'meta' : 'IMPUT_SAS_SPSS/INPUT_SAS_ENEM_2001.sas',\n",
    "         'data' : 'DADOS/DADOS_ENEM_2001.TXT', \n",
    "         'rename' : 'DADOS_ENEM_2001.csv'},\n",
    "    'micro_enem2002.zip' : \n",
    "        {'meta' : 'INPUTS_SAS_SPSS/INPUT_SAS_ENEM_2002.sas',\n",
    "         'data' : 'DADOS/DADOS_ENEM_2002.zip', \n",
    "         'rename' : 'DADOS_ENEM_2002.csv'},\n",
    "    'micro_enem2003.zip' : \n",
    "        {'meta' : 'INPUTS_SAS_SPSS/INPUT_SAS_ENEM_2003.sas',\n",
    "         'data' : 'DADOS/DADOS_ENEM_2003.zip', \n",
    "         'rename' : 'DADOS_ENEM_2003.csv'},\n",
    "    'micro_enem2004.zip' : \n",
    "        {'meta' : 'INPUTS_SAS_SPSS/INPUT_SAS_ENEM_2004.sas',\n",
    "         'data' : 'DADOS/DADOS_ENEM_2004.zip', \n",
    "         'rename' : 'DADOS_ENEM_2004.csv'},\n",
    "    'microdados_enem_2005.zip' : \n",
    "        {'meta' : 'INPUT_SAS_ENEM_2005.sas',\n",
    "         'data' : 'DADOS_ENEM_2005.zip', \n",
    "         'rename' : 'DADOS_ENEM_2005.csv'},\n",
    "    'microdados_ENEM_2006.zip' : \n",
    "        {'meta' : '',\n",
    "         'data' : 'Microdados Enem 2006 (2° Ediç╞o)/Dados/MICRODADOS_ENEM_2006.csv', \n",
    "         'rename' : 'DADOS_ENEM_2006.csv'},\n",
    "    'microdados_enem_2007_DVD.zip' : \n",
    "        {'meta' : '',\n",
    "         'data' : 'Microdados Enem 2007 (2° Ediç╞o)/Dados/MICRODADOS_ENEM_2007.csv', \n",
    "         'rename' : 'DADOS_ENEM_2007.csv'},\n",
    "    'microdados_enem_2008.zip' : \n",
    "        {'meta' : '',\n",
    "         'data' : 'Microdados Enem 2008 (2° Ediç╞o)/Dados/MICRODADOS_ENEM_2008.csv', \n",
    "         'rename' : 'DADOS_ENEM_2008.csv'}\n",
    "}\n",
    "\n",
    "def transform_sas_to_csv(sas_filename):\n",
    "\n",
    "    dir_name = os.path.split(sas_filename)[0]\n",
    "    csv_filename = \"\"\n",
    "    \n",
    "    pre_fields = collections.OrderedDict()\n",
    "\n",
    "    # get info from malformed sas file\n",
    "    try:\n",
    "        sas_file = open(sas_filename, \"r\", encoding=\"ISO-8859-1\")\n",
    "    except:\n",
    "        sys.exit(\"File not found\")\n",
    "    for line in sas_file:\n",
    "        match = re.search('^INFILE.*', line)\n",
    "        if match:\n",
    "            data_filename = os.path.join(dir_name, line.split('\"')[1].split('\\\\')[-1])\n",
    "            csv_filename = \"{}.csv\".format(data_filename)\n",
    "        match = re.search('^@([0-9]+)[ \\t]+([^ \\t]+)', line)\n",
    "        if match:\n",
    "            pre_fields[match.group(2)] = int(match.group(1))-1\n",
    "    pre_fields['ending'] = None\n",
    "    sas_file.close()\n",
    "\n",
    "    # post-processing fields\n",
    "    pre_value = None\n",
    "    pre_key = None\n",
    "    fields = collections.OrderedDict()\n",
    "    for key, value in pre_fields.items():\n",
    "        if pre_key:\n",
    "            fields[pre_key] = [pre_value, value]\n",
    "        pre_value = value\n",
    "        pre_key = key\n",
    "\n",
    "    # open csv file to write\n",
    "    csv_file = open(csv_filename, mode='w')\n",
    "    fieldnames = list(fields.keys())\n",
    "    writer = csv.DictWriter(csv_file, fieldnames=fieldnames, delimiter=\";\")\n",
    "    writer.writeheader()\n",
    "\n",
    "    # open data file\n",
    "    try:\n",
    "        data_file = open(data_filename, \"r\", encoding=\"ISO-8859-1\")\n",
    "        out.value = \"Data file opened\"\n",
    "    except:\n",
    "        return \"\"\n",
    "    n = 1\n",
    "    for line in data_file:\n",
    "        # read data file line by line\n",
    "        csv_line = dict()\n",
    "        for k, v in fields.items():\n",
    "            csv_line[k] = str.strip(line[v[0]: v[1]])\n",
    "        writer.writerow(csv_line)\n",
    "        if n % 1000 == 0:\n",
    "            out.value = \"Reading line {}\".format(n)\n",
    "        n += 1\n",
    "    \n",
    "    csv_file.close()\n",
    "    data_file.close()\n",
    "    return csv_filename\n",
    "    \n",
    "\n",
    "for zipname, zipcompressedfiles in files.items():\n",
    "    data = zipcompressedfiles['data']\n",
    "    meta = zipcompressedfiles['meta']\n",
    "    rename = zipcompressedfiles['rename']\n",
    "    \n",
    "    out.value = \"Downloading {}\".format(zipname)\n",
    "    enem = requests.get(\"{}{}\".format(urlprefix,zipname))\n",
    "    z = zipfile.ZipFile(io.BytesIO(enem.content))\n",
    "    z.extract(data, path='tmp')\n",
    "    \n",
    "    if meta:\n",
    "        z.extract(meta, path='tmp')\n",
    "        if '.zip' in data:\n",
    "            z = zipfile.ZipFile(os.path.join('tmp', data))\n",
    "            z.extractall(path='tmp')\n",
    "            data = z.namelist()[0]\n",
    "            \n",
    "        os.rename(os.path.join('tmp', data), os.path.join('raw', os.path.basename(data)))\n",
    "        os.rename(os.path.join('tmp', meta), os.path.join('raw', os.path.basename(meta)))\n",
    "        csv_name = transform_sas_to_csv(os.path.join('raw', os.path.basename(meta)))\n",
    "        os.rename(csv_name, os.path.join('data', rename))\n",
    "    else:\n",
    "        os.rename(os.path.join('tmp', data), os.path.join('data', rename))\n",
    "    out.value = \"{} finished.\".format(os.path.basename(data))\n",
    "out.value = \"All data downloaded and processed.\"\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then, it is necessary to load the data dictionary, already described above:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>YEAR</th>\n",
       "      <th>FIELD</th>\n",
       "      <th>TYPE</th>\n",
       "      <th>RENAME_TO</th>\n",
       "      <th>MANDATORY</th>\n",
       "      <th>CONVERT_TO</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>1999</td>\n",
       "      <td>Q3</td>\n",
       "      <td>category</td>\n",
       "      <td>SKIN_COLOR</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'A': 'white', 'B': 'biracial', 'C': 'black', ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>1999</td>\n",
       "      <td>Q18</td>\n",
       "      <td>category</td>\n",
       "      <td>FATHER_SCHOOLING</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'A': 'A-didnt_go_to_school', 'B': 'B-primary_...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>1999</td>\n",
       "      <td>Q19</td>\n",
       "      <td>category</td>\n",
       "      <td>MOTHER_SCHOOLING</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'A': 'A-didnt_go_to_school', 'B': 'B-primary_...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>1999</td>\n",
       "      <td>Q55</td>\n",
       "      <td>category</td>\n",
       "      <td>SECONDARY_SCHOOL_1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>{'A': 'daytime', 'B': 'both', 'C': 'nighttime'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>1999</td>\n",
       "      <td>Q56</td>\n",
       "      <td>category</td>\n",
       "      <td>SECONDARY_SCHOOL_2</td>\n",
       "      <td>1.0</td>\n",
       "      <td>{'A': 'public', 'B': 'both', 'C': 'private'}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    YEAR FIELD      TYPE           RENAME_TO  MANDATORY  \\\n",
       "19  1999    Q3  category          SKIN_COLOR        NaN   \n",
       "23  1999   Q18  category    FATHER_SCHOOLING        NaN   \n",
       "24  1999   Q19  category    MOTHER_SCHOOLING        NaN   \n",
       "26  1999   Q55  category  SECONDARY_SCHOOL_1        1.0   \n",
       "27  1999   Q56  category  SECONDARY_SCHOOL_2        1.0   \n",
       "\n",
       "                                           CONVERT_TO  \n",
       "19  {'A': 'white', 'B': 'biracial', 'C': 'black', ...  \n",
       "23  {'A': 'A-didnt_go_to_school', 'B': 'B-primary_...  \n",
       "24  {'A': 'A-didnt_go_to_school', 'B': 'B-primary_...  \n",
       "26    {'A': 'daytime', 'B': 'both', 'C': 'nighttime'}  \n",
       "27       {'A': 'public', 'B': 'both', 'C': 'private'}  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "dicts = pd.read_csv(\"table_of_dicts_enem.csv\", delimiter='\\t')\n",
    "dicts.dropna(subset=['CONVERT_TO']).head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Loading data (optional step)\n",
    "\n",
    "The data was uploaded to a single DataFrame. To facilitate analysis on a conventional computer, a maximum of 100000 rows per year were considered. Obviously, if you have a computer with a lot of memory, we encourage you to raise that level to 200,000 or more, but we believe that the statistical effect will not be so impacting in the analysis.\n",
    "\n",
    "One important thing to note is that the encoding of files is different. Files from 1999 to 2005 are encoded in `utf-8` whereas those from 2006 to 2008 are encoded in` latin1`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "79c8b6fc0a9042d09fdce67d501e2794",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "HTML(value='', placeholder='')"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import ast\n",
    "import os\n",
    "import time\n",
    "from IPython.display import display\n",
    "import ipywidgets as widgets\n",
    "\n",
    "out = widgets.HTML(value='',\n",
    "    placeholder='',\n",
    "    description='')\n",
    "display(out)\n",
    "\n",
    "start_time = time.time()\n",
    "\n",
    "enem = pd.DataFrame()\n",
    "\n",
    "files = {\n",
    "    1999 : 'utf-8', 2000 : 'utf-8', 2001 : 'utf-8', 2002 : 'utf-8', 2003 : 'utf-8', \n",
    "    2004 : 'utf-8', 2005 : 'utf-8', 2006 : 'latin1', 2007 : 'latin1', 2008 : 'latin1' \n",
    "}\n",
    "\n",
    "for year, enc in files.items():\n",
    "    # Which columns we would like to load for this year\n",
    "    columns = list(dicts.loc[dicts['YEAR']==year , 'FIELD'])\n",
    "    # Which types are they?\n",
    "    types = dicts.loc[dicts['YEAR']==year, ['FIELD', 'TYPE']].\\\n",
    "        dropna().set_index('FIELD').squeeze().to_dict()\n",
    "    # Only non-NaN fields:\n",
    "    mandatory = list(dicts.loc[(dicts['YEAR']==year) & (dicts['MANDATORY']==1) , 'FIELD'])\n",
    "    out.value = \"Loading DADOS_ENEM_{}.csv...\".format(year)\n",
    "    \n",
    "    enem_year = pd.read_csv(os.path.join(\"data\", 'DADOS_ENEM_{}.csv'.format(year)), \n",
    "        delimiter=';', \n",
    "        usecols=columns,\n",
    "        dtype=types,\n",
    "        encoding=enc\n",
    "    ).dropna(\n",
    "        subset=mandatory\n",
    "    )\n",
    "    \n",
    "    # Which will be the new names of the fields?\n",
    "    new_names = dicts.loc[dicts['YEAR']==year, ['FIELD', 'RENAME_TO']].\\\n",
    "        dropna().set_index('FIELD').squeeze().to_dict()\n",
    "    \n",
    "    enem_year.rename(\n",
    "        index=str, \n",
    "        columns=new_names, \n",
    "        inplace=True)\n",
    "    \n",
    "    # Sampling down to 100000\n",
    "    if len(enem_year) > 100000:\n",
    "        enem_year = enem_year.sample(n = 100000)\n",
    "    \n",
    "    # Changing fields contents:\n",
    "    new_values = dicts.loc[dicts['YEAR']==year, ['RENAME_TO', 'CONVERT_TO']].dropna()\n",
    "    \n",
    "    for index, row in new_values.iterrows():\n",
    "        # Converting string representation of a dict to a real dict (be careful and \n",
    "        #   check csv file before running!)\n",
    "        d = ast.literal_eval(row[1])\n",
    "        enem_year[row[0]].replace(d.keys(), d.values(), inplace=True)\n",
    "        \n",
    "    # Putting all together\n",
    "    enem = pd.concat([enem, enem_year], sort=False)\n",
    "    del enem_year\n",
    "\n",
    "out.value = \"All data loaded in {} seconds.\".format(round(time.time() - start_time, 2))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before go on, let's enforce data types:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# First we make a copy of the original field names, new names and their desired types:\n",
    "ct = dicts[['FIELD', 'RENAME_TO', 'TYPE']].copy()\n",
    "# If RENAME_TO is empty, we fill with original field name:\n",
    "ct['RENAME_TO'].fillna(ct['FIELD'], inplace=True)\n",
    "# Now FIELD needs to be dropped, and RENAME_TO will be renamed to FIELD\n",
    "ct.drop(columns='FIELD', inplace=True)\n",
    "ct.columns = ['FIELD', 'TYPE']\n",
    "# We just drop the rows that have NaN values.\n",
    "# These field types will be guessed at load time and their types are ok:\n",
    "ct.dropna(inplace=True)\n",
    "# We set the index \n",
    "ct.set_index('FIELD', inplace=True)\n",
    "for index, row in ct.iterrows():\n",
    "    if index in enem.columns:\n",
    "        enem[index] = enem[index].astype(row[0])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And save DataFrame in an .hdf5 file in order to reload this data as fast as possible:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- 93.55 seconds to save ---\n"
     ]
    }
   ],
   "source": [
    "start_time = time.time()\n",
    "\n",
    "s = pd.HDFStore('enem.h5')\n",
    "s.put('enem', enem, format='table', data_columns=True)\n",
    "s.close()\n",
    "\n",
    "print(\"--- {} seconds to save ---\".format(round(time.time() - start_time, 2)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fast data loading\n",
    "\n",
    "The entire previous section can be skipped if you just want to access the results. In the file `enem.h5` are the data of up to 100000 students per year, from the years 1999 to 2008, chosen randomly. Thus, it is possible to skip the slow portion of data load and leave only the part of data analysis and visualization.\n",
    "To load the data, just enter the following commands:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--- 22.55 seconds to reload ---\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import time\n",
    "start_time = time.time()\n",
    "\n",
    "enem = pd.read_hdf('enem.h5', 'enem')\n",
    "\n",
    "print(\"--- {} seconds to reload ---\".format(round(time.time() - start_time, 2)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Data will be loaded from the medium directly into the DataFrame. In tests done on a typical computer, comparison between the load speed of the original .csv files and the .h5 file was 80 to 100 times faster."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data analysis\n",
    "\n",
    "### Difference of grades between public and private schools\n",
    "\n",
    "Due to profound social problems in Brazil, we think the first interesting analysis is to observe the difference of grades between public and private schools, as well as the standard deviations of these two schools. Let's look at the differences between multiple choice test and essay scores.\n",
    "\n",
    "First, from the multiple choice test:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "ppo_mean = enem.groupby(['YEAR', 'SECONDARY_SCHOOL_2']).mean()['MC_SCORE']\n",
    "ppo_mean_public = ppo_mean.loc[ppo_mean.index.get_level_values('SECONDARY_SCHOOL_2') == 'public'].xs('public', level=1)\n",
    "ppo_mean_private = ppo_mean.loc[ppo_mean.index.get_level_values('SECONDARY_SCHOOL_2') == 'private'].xs('private', level=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "ppo_std = enem.groupby(['YEAR', 'SECONDARY_SCHOOL_2']).std()['MC_SCORE']\n",
    "ppo_std_public = ppo_std.loc[ppo_std.index.get_level_values('SECONDARY_SCHOOL_2') == 'public'].xs('public', level=1)\n",
    "ppo_std_private = ppo_std.loc[ppo_std.index.get_level_values('SECONDARY_SCHOOL_2') == 'private'].xs('private', level=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "By plotting the averages and standard deviations of students in public and private schools, we have:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>MEAN_PUBLIC</th>\n",
       "      <th>MEAN_PRIVATE</th>\n",
       "      <th>STD_PUBLIC</th>\n",
       "      <th>STD_PRIVATE</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YEAR</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>43.421078</td>\n",
       "      <td>54.611563</td>\n",
       "      <td>16.362298</td>\n",
       "      <td>17.429100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>48.545539</td>\n",
       "      <td>57.003091</td>\n",
       "      <td>18.190014</td>\n",
       "      <td>18.072709</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>36.144157</td>\n",
       "      <td>47.681297</td>\n",
       "      <td>12.326296</td>\n",
       "      <td>15.975449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2002</th>\n",
       "      <td>29.816776</td>\n",
       "      <td>41.873310</td>\n",
       "      <td>10.599267</td>\n",
       "      <td>15.658384</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>43.673948</td>\n",
       "      <td>59.447216</td>\n",
       "      <td>17.280009</td>\n",
       "      <td>19.323924</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004</th>\n",
       "      <td>38.031025</td>\n",
       "      <td>55.781753</td>\n",
       "      <td>15.299977</td>\n",
       "      <td>20.014123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>36.046699</td>\n",
       "      <td>47.931261</td>\n",
       "      <td>13.728776</td>\n",
       "      <td>18.634165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2006</th>\n",
       "      <td>34.922681</td>\n",
       "      <td>50.577074</td>\n",
       "      <td>11.571798</td>\n",
       "      <td>16.533123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>47.869393</td>\n",
       "      <td>65.006827</td>\n",
       "      <td>15.321747</td>\n",
       "      <td>17.975047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008</th>\n",
       "      <td>39.477474</td>\n",
       "      <td>56.861266</td>\n",
       "      <td>13.376375</td>\n",
       "      <td>16.755832</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      MEAN_PUBLIC  MEAN_PRIVATE  STD_PUBLIC  STD_PRIVATE\n",
       "YEAR                                                    \n",
       "1999    43.421078     54.611563   16.362298    17.429100\n",
       "2000    48.545539     57.003091   18.190014    18.072709\n",
       "2001    36.144157     47.681297   12.326296    15.975449\n",
       "2002    29.816776     41.873310   10.599267    15.658384\n",
       "2003    43.673948     59.447216   17.280009    19.323924\n",
       "2004    38.031025     55.781753   15.299977    20.014123\n",
       "2005    36.046699     47.931261   13.728776    18.634165\n",
       "2006    34.922681     50.577074   11.571798    16.533123\n",
       "2007    47.869393     65.006827   15.321747    17.975047\n",
       "2008    39.477474     56.861266   13.376375    16.755832"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_std_o = pd.DataFrame(\n",
    "    {'MEAN_PUBLIC': ppo_mean_public, 'MEAN_PRIVATE':ppo_mean_private, \n",
    "     'STD_PUBLIC': ppo_std_public, 'STD_PRIVATE':ppo_std_private}\n",
    ")\n",
    "mean_std_o"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f3d4cf073c8>"
      ]
     },
     "execution_count": 8,
     "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": [
    "import matplotlib.pyplot as plt \n",
    "%matplotlib inline\n",
    "\n",
    "mean_std_o.reset_index(inplace=True)\n",
    "mean_std_o['YEAR'] = mean_std_o['YEAR'].astype(int)\n",
    "plt.plot('YEAR', 'MEAN_PUBLIC', data=mean_std_o, marker='', color='blue', linewidth=2)\n",
    "plt.plot('YEAR', 'MEAN_PRIVATE', data=mean_std_o, marker='', color='blue', linewidth=2, linestyle='dashed')\n",
    "plt.plot('YEAR', 'STD_PUBLIC', data=mean_std_o, marker='', color='olive', linewidth=2)\n",
    "plt.plot('YEAR', 'STD_PRIVATE', data=mean_std_o, marker='', color='olive', linewidth=2, linestyle='dashed')\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can observe in this analysis that the scores of the public and private schools have the same \"drawing\" and are spaced in a more or less constant form. To check this phenomenon, let's construct a column of difference between the notes and plot your chart:\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x7f3d4cfc75f8>]"
      ]
     },
     "execution_count": 9,
     "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": [
    "mean_std_o['DIFF'] = mean_std_o['MEAN_PRIVATE'] - mean_std_o['MEAN_PUBLIC']\n",
    "plt.plot('YEAR', 'DIFF', data=mean_std_o, marker='', color='blue', linewidth=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "17.75072733327714"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_std_o['DIFF'].max()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8.457552896753057"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_std_o['DIFF'].min()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, let's do the same to essay scores:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "ppe_mean = enem.groupby(['YEAR', 'SECONDARY_SCHOOL_2']).mean()['ESSAY_SCORE']\n",
    "ppe_mean_public = ppe_mean.loc[ppe_mean.index.get_level_values('SECONDARY_SCHOOL_2') == 'public'].xs('public', level=1)\n",
    "ppe_mean_private = ppe_mean.loc[ppe_mean.index.get_level_values('SECONDARY_SCHOOL_2') == 'private'].xs('private', level=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "ppe_std = enem.groupby(['YEAR', 'SECONDARY_SCHOOL_2']).std()['ESSAY_SCORE']\n",
    "ppe_std_public = ppe_std.loc[ppe_std.index.get_level_values('SECONDARY_SCHOOL_2') == 'public'].xs('public', level=1)\n",
    "ppe_std_private = ppe_std.loc[ppe_std.index.get_level_values('SECONDARY_SCHOOL_2') == 'private'].xs('private', level=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>MEAN_PUBLIC</th>\n",
       "      <th>MEAN_PRIVATE</th>\n",
       "      <th>STD_PUBLIC</th>\n",
       "      <th>STD_PRIVATE</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YEAR</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>41.852936</td>\n",
       "      <td>48.580293</td>\n",
       "      <td>22.561681</td>\n",
       "      <td>23.061552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>54.001700</td>\n",
       "      <td>55.296372</td>\n",
       "      <td>21.649851</td>\n",
       "      <td>23.545837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>47.766693</td>\n",
       "      <td>53.629413</td>\n",
       "      <td>17.814134</td>\n",
       "      <td>20.693039</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2002</th>\n",
       "      <td>49.055042</td>\n",
       "      <td>54.059500</td>\n",
       "      <td>17.958081</td>\n",
       "      <td>21.969717</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>47.585288</td>\n",
       "      <td>51.420263</td>\n",
       "      <td>20.764072</td>\n",
       "      <td>25.985926</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004</th>\n",
       "      <td>36.974702</td>\n",
       "      <td>48.505747</td>\n",
       "      <td>25.484605</td>\n",
       "      <td>27.964092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>52.939522</td>\n",
       "      <td>53.942982</td>\n",
       "      <td>17.537182</td>\n",
       "      <td>22.814929</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2006</th>\n",
       "      <td>50.389122</td>\n",
       "      <td>54.215586</td>\n",
       "      <td>19.040550</td>\n",
       "      <td>22.969008</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>53.769430</td>\n",
       "      <td>60.500689</td>\n",
       "      <td>15.593211</td>\n",
       "      <td>16.558895</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008</th>\n",
       "      <td>57.677200</td>\n",
       "      <td>62.530354</td>\n",
       "      <td>14.822475</td>\n",
       "      <td>18.419045</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      MEAN_PUBLIC  MEAN_PRIVATE  STD_PUBLIC  STD_PRIVATE\n",
       "YEAR                                                    \n",
       "1999    41.852936     48.580293   22.561681    23.061552\n",
       "2000    54.001700     55.296372   21.649851    23.545837\n",
       "2001    47.766693     53.629413   17.814134    20.693039\n",
       "2002    49.055042     54.059500   17.958081    21.969717\n",
       "2003    47.585288     51.420263   20.764072    25.985926\n",
       "2004    36.974702     48.505747   25.484605    27.964092\n",
       "2005    52.939522     53.942982   17.537182    22.814929\n",
       "2006    50.389122     54.215586   19.040550    22.969008\n",
       "2007    53.769430     60.500689   15.593211    16.558895\n",
       "2008    57.677200     62.530354   14.822475    18.419045"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mean_std_e = pd.DataFrame(\n",
    "    {'MEAN_PUBLIC': ppe_mean_public, 'MEAN_PRIVATE':ppe_mean_private, \n",
    "     'STD_PUBLIC': ppe_std_public, 'STD_PRIVATE':ppe_std_private}\n",
    ")\n",
    "mean_std_e"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f3d4d06a0b8>"
      ]
     },
     "execution_count": 15,
     "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": [
    "mean_std_e.reset_index(inplace=True)\n",
    "mean_std_e['YEAR'] = mean_std_e['YEAR'].astype(int)\n",
    "plt.plot('YEAR', 'MEAN_PUBLIC', data=mean_std_e, marker='', color='blue', linewidth=2)\n",
    "plt.plot('YEAR', 'MEAN_PRIVATE', data=mean_std_e, marker='', color='blue', linewidth=2, linestyle='dashed')\n",
    "plt.plot('YEAR', 'STD_PUBLIC', data=mean_std_e, marker='', color='olive', linewidth=2)\n",
    "plt.plot('YEAR', 'STD_PRIVATE', data=mean_std_e, marker='', color='olive', linewidth=2, linestyle='dashed')\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It was noted that the behavior of the private school grades was higher, but there were large differences between the differences in school grades, noting the great difference in 2004, with a high standard deviation, both for public schools and individuals."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Difference of notes regarding infrastructure and basic sanitation\n",
    "\n",
    "Let's compare now, using the same data, two different extracts: students living in homes with tap water and electricity and students who do not have tap water and electricity at home.\n",
    "\n",
    "First let's bring together all the students who have both resources:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "we_mean = enem.groupby(['YEAR', 'TAP_WATER', 'ELECTRIFIED_HOME']).mean()['MC_SCORE']\n",
    "we_mean_both = we_mean.loc[\n",
    "    (we_mean.index.get_level_values('TAP_WATER') == 'yes') & \n",
    "    (we_mean.index.get_level_values('ELECTRIFIED_HOME') == 'yes')\n",
    "].xs('yes', level=2).xs('yes', level=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "we_mean_none = we_mean.loc[\n",
    "    (we_mean.index.get_level_values('TAP_WATER') == 'no') & \n",
    "    (we_mean.index.get_level_values('ELECTRIFIED_HOME') == 'no')\n",
    "].xs('no', level=2).xs('no', level=1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>MEAN_BOTH</th>\n",
       "      <th>MEAN_NONE</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YEAR</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>47.110741</td>\n",
       "      <td>44.661204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>51.184570</td>\n",
       "      <td>49.398145</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>37.817669</td>\n",
       "      <td>32.638472</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2002</th>\n",
       "      <td>31.409227</td>\n",
       "      <td>26.360778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>46.227707</td>\n",
       "      <td>39.414856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004</th>\n",
       "      <td>40.470997</td>\n",
       "      <td>34.972897</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>37.043010</td>\n",
       "      <td>30.969300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2006</th>\n",
       "      <td>37.274898</td>\n",
       "      <td>30.088836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>49.812652</td>\n",
       "      <td>41.459613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008</th>\n",
       "      <td>42.317030</td>\n",
       "      <td>35.713972</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      MEAN_BOTH  MEAN_NONE\n",
       "YEAR                      \n",
       "1999  47.110741  44.661204\n",
       "2000  51.184570  49.398145\n",
       "2001  37.817669  32.638472\n",
       "2002  31.409227  26.360778\n",
       "2003  46.227707  39.414856\n",
       "2004  40.470997  34.972897\n",
       "2005  37.043010  30.969300\n",
       "2006  37.274898  30.088836\n",
       "2007  49.812652  41.459613\n",
       "2008  42.317030  35.713972"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "we_mean_df = pd.DataFrame(\n",
    "    {'MEAN_BOTH': we_mean_both, 'MEAN_NONE':we_mean_none}\n",
    ")\n",
    "we_mean_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f3d4d12a0b8>"
      ]
     },
     "execution_count": 19,
     "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": [
    "we_mean_df.reset_index(inplace=True)\n",
    "we_mean_df['YEAR'] = we_mean_df['YEAR'].astype(int)\n",
    "plt.plot('YEAR', 'MEAN_BOTH', data=we_mean_df, marker='', color='blue', linewidth=2)\n",
    "plt.plot('YEAR', 'MEAN_NONE', data=we_mean_df, marker='', color='olive', linewidth=2)\n",
    "\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Difference in relation to total family income\n",
    "\n",
    "Another interesting analysis is to measure the average grade as a function of total family income. The total monthly family income was organized into 8 bands: 1 minimum wage or less; 1 to 2 minimum wages; 2 to 5 minimum wages; 5 to 10 minimum wages; 10 to 30 minimum wages; 30 to 50 minimum wages; more than 50 minimum wages; and no monthly income:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>A_1_MW_or_less</th>\n",
       "      <th>B_1_to_2_MW</th>\n",
       "      <th>C_2_to_5_MW</th>\n",
       "      <th>D_5_to_10_MW</th>\n",
       "      <th>E_10_to_30_MW</th>\n",
       "      <th>F_30_to_50_MW</th>\n",
       "      <th>G_50+_MW</th>\n",
       "      <th>H_no_income</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YEAR</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>33.968140</td>\n",
       "      <td>36.554268</td>\n",
       "      <td>41.410276</td>\n",
       "      <td>47.226818</td>\n",
       "      <td>53.802565</td>\n",
       "      <td>57.881187</td>\n",
       "      <td>58.774203</td>\n",
       "      <td>44.168856</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>36.457261</td>\n",
       "      <td>38.041872</td>\n",
       "      <td>45.433937</td>\n",
       "      <td>51.942112</td>\n",
       "      <td>57.945445</td>\n",
       "      <td>61.018604</td>\n",
       "      <td>59.859634</td>\n",
       "      <td>45.967342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>30.640216</td>\n",
       "      <td>33.341829</td>\n",
       "      <td>37.314053</td>\n",
       "      <td>41.008144</td>\n",
       "      <td>46.642202</td>\n",
       "      <td>50.436142</td>\n",
       "      <td>50.938898</td>\n",
       "      <td>32.347530</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2002</th>\n",
       "      <td>25.951000</td>\n",
       "      <td>27.887710</td>\n",
       "      <td>31.288648</td>\n",
       "      <td>34.886380</td>\n",
       "      <td>40.368790</td>\n",
       "      <td>43.573502</td>\n",
       "      <td>44.366829</td>\n",
       "      <td>26.640386</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>34.953358</td>\n",
       "      <td>40.429991</td>\n",
       "      <td>48.886515</td>\n",
       "      <td>55.077630</td>\n",
       "      <td>61.693604</td>\n",
       "      <td>63.260614</td>\n",
       "      <td>57.220432</td>\n",
       "      <td>35.175571</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004</th>\n",
       "      <td>31.229182</td>\n",
       "      <td>35.502523</td>\n",
       "      <td>42.161082</td>\n",
       "      <td>47.785604</td>\n",
       "      <td>54.412509</td>\n",
       "      <td>54.241201</td>\n",
       "      <td>50.907652</td>\n",
       "      <td>32.105335</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>30.073298</td>\n",
       "      <td>34.296858</td>\n",
       "      <td>39.887689</td>\n",
       "      <td>44.139720</td>\n",
       "      <td>50.406092</td>\n",
       "      <td>47.345833</td>\n",
       "      <td>46.940305</td>\n",
       "      <td>28.725878</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2006</th>\n",
       "      <td>30.929945</td>\n",
       "      <td>34.203323</td>\n",
       "      <td>38.757741</td>\n",
       "      <td>44.976785</td>\n",
       "      <td>54.519120</td>\n",
       "      <td>57.727555</td>\n",
       "      <td>52.297337</td>\n",
       "      <td>30.517723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>42.222302</td>\n",
       "      <td>47.263009</td>\n",
       "      <td>53.712815</td>\n",
       "      <td>59.654524</td>\n",
       "      <td>69.200712</td>\n",
       "      <td>71.510000</td>\n",
       "      <td>59.477701</td>\n",
       "      <td>39.718785</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008</th>\n",
       "      <td>34.166731</td>\n",
       "      <td>38.455831</td>\n",
       "      <td>44.678873</td>\n",
       "      <td>52.461760</td>\n",
       "      <td>61.050033</td>\n",
       "      <td>64.236391</td>\n",
       "      <td>57.220257</td>\n",
       "      <td>32.648478</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      A_1_MW_or_less  B_1_to_2_MW  C_2_to_5_MW  D_5_to_10_MW  E_10_to_30_MW  \\\n",
       "YEAR                                                                          \n",
       "1999       33.968140    36.554268    41.410276     47.226818      53.802565   \n",
       "2000       36.457261    38.041872    45.433937     51.942112      57.945445   \n",
       "2001       30.640216    33.341829    37.314053     41.008144      46.642202   \n",
       "2002       25.951000    27.887710    31.288648     34.886380      40.368790   \n",
       "2003       34.953358    40.429991    48.886515     55.077630      61.693604   \n",
       "2004       31.229182    35.502523    42.161082     47.785604      54.412509   \n",
       "2005       30.073298    34.296858    39.887689     44.139720      50.406092   \n",
       "2006       30.929945    34.203323    38.757741     44.976785      54.519120   \n",
       "2007       42.222302    47.263009    53.712815     59.654524      69.200712   \n",
       "2008       34.166731    38.455831    44.678873     52.461760      61.050033   \n",
       "\n",
       "      F_30_to_50_MW   G_50+_MW  H_no_income  \n",
       "YEAR                                         \n",
       "1999      57.881187  58.774203    44.168856  \n",
       "2000      61.018604  59.859634    45.967342  \n",
       "2001      50.436142  50.938898    32.347530  \n",
       "2002      43.573502  44.366829    26.640386  \n",
       "2003      63.260614  57.220432    35.175571  \n",
       "2004      54.241201  50.907652    32.105335  \n",
       "2005      47.345833  46.940305    28.725878  \n",
       "2006      57.727555  52.297337    30.517723  \n",
       "2007      71.510000  59.477701    39.718785  \n",
       "2008      64.236391  57.220257    32.648478  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inc_mean = enem.groupby(['YEAR', 'FAMILY_INCOME']).mean()['MC_SCORE']\n",
    "inc_mean_A = inc_mean.loc[inc_mean.index.get_level_values('FAMILY_INCOME') == 'A_1_MW_or_less'].xs('A_1_MW_or_less', level=1)\n",
    "inc_mean_B = inc_mean.loc[inc_mean.index.get_level_values('FAMILY_INCOME') == 'B_1_to_2_MW'].xs('B_1_to_2_MW', level=1)\n",
    "inc_mean_C = inc_mean.loc[inc_mean.index.get_level_values('FAMILY_INCOME') == 'C_2_to_5_MW'].xs('C_2_to_5_MW', level=1)\n",
    "inc_mean_D = inc_mean.loc[inc_mean.index.get_level_values('FAMILY_INCOME') == 'D_5_to_10_MW'].xs('D_5_to_10_MW', level=1)\n",
    "inc_mean_E = inc_mean.loc[inc_mean.index.get_level_values('FAMILY_INCOME') == 'E_10_to_30_MW'].xs('E_10_to_30_MW', level=1)\n",
    "inc_mean_F = inc_mean.loc[inc_mean.index.get_level_values('FAMILY_INCOME') == 'F_30_to_50_MW'].xs('F_30_to_50_MW', level=1)\n",
    "inc_mean_G = inc_mean.loc[inc_mean.index.get_level_values('FAMILY_INCOME') == 'G_50+_MW'].xs('G_50+_MW', level=1)\n",
    "inc_mean_H = inc_mean.loc[inc_mean.index.get_level_values('FAMILY_INCOME') == 'H_no_income'].xs('H_no_income', level=1)\n",
    "inc_mean_df = pd.DataFrame(\n",
    "    {'A_1_MW_or_less': inc_mean_A, 'B_1_to_2_MW':inc_mean_B, \n",
    "     'C_2_to_5_MW': inc_mean_C, 'D_5_to_10_MW':inc_mean_D, \n",
    "     'E_10_to_30_MW': inc_mean_E, 'F_30_to_50_MW':inc_mean_F, \n",
    "     'G_50+_MW': inc_mean_G, 'H_no_income':inc_mean_H}\n",
    ")\n",
    "inc_mean_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f3d4d245a58>"
      ]
     },
     "execution_count": 21,
     "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": [
    "df_tmp = inc_mean_df.reset_index()\n",
    "df_tmp['YEAR'] = df_tmp['YEAR'].astype(int)\n",
    "\n",
    "plt.plot('YEAR', 'A_1_MW_or_less', data=df_tmp, marker='', color='blue', linewidth=2)\n",
    "plt.plot('YEAR', 'B_1_to_2_MW', data=df_tmp, marker='', color='olive', linewidth=2)\n",
    "plt.plot('YEAR', 'C_2_to_5_MW', data=df_tmp, marker='', color='green', linewidth=2)\n",
    "plt.plot('YEAR', 'D_5_to_10_MW', data=df_tmp, marker='', color='red', linewidth=2)\n",
    "plt.plot('YEAR', 'E_10_to_30_MW', data=df_tmp, marker='', color='cyan', linewidth=2)\n",
    "plt.plot('YEAR', 'F_30_to_50_MW', data=df_tmp, marker='', color='magenta', linewidth=2)\n",
    "plt.plot('YEAR', 'G_50+_MW', data=df_tmp, marker='', color='navy', linewidth=2)\n",
    "plt.plot('YEAR', 'H_no_income', data=df_tmp, marker='', color='black', linewidth=2)\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Difference regarding preparatory course\n",
    "Another aspect to consider is to verify if the candidate attended a preparatory course in the year of the exam:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "pc_mean = enem.groupby(['YEAR', 'CURRENTY_PREP_COURSE']).mean()['MC_SCORE']\n",
    "pc_mean_yes = pc_mean.loc[\n",
    "    pc_mean.index.get_level_values('CURRENTY_PREP_COURSE') == 'yes'\n",
    "].xs('yes', level=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "metadata": {},
   "outputs": [],
   "source": [
    "pc_mean_no = pc_mean.loc[\n",
    "    pc_mean.index.get_level_values('CURRENTY_PREP_COURSE') == 'no'\n",
    "].xs('no', level=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "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>yes</th>\n",
       "      <th>no</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YEAR</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1999</th>\n",
       "      <td>45.997518</td>\n",
       "      <td>44.521377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000</th>\n",
       "      <td>53.784133</td>\n",
       "      <td>43.348239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <td>37.014521</td>\n",
       "      <td>39.725550</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2002</th>\n",
       "      <td>35.805411</td>\n",
       "      <td>30.407299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2003</th>\n",
       "      <td>52.035079</td>\n",
       "      <td>41.015679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2004</th>\n",
       "      <td>47.733150</td>\n",
       "      <td>36.108457</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>40.516131</td>\n",
       "      <td>35.455183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2006</th>\n",
       "      <td>42.039371</td>\n",
       "      <td>35.327378</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2007</th>\n",
       "      <td>51.622581</td>\n",
       "      <td>48.243659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2008</th>\n",
       "      <td>46.994654</td>\n",
       "      <td>40.334809</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            yes         no\n",
       "YEAR                      \n",
       "1999  45.997518  44.521377\n",
       "2000  53.784133  43.348239\n",
       "2001  37.014521  39.725550\n",
       "2002  35.805411  30.407299\n",
       "2003  52.035079  41.015679\n",
       "2004  47.733150  36.108457\n",
       "2005  40.516131  35.455183\n",
       "2006  42.039371  35.327378\n",
       "2007  51.622581  48.243659\n",
       "2008  46.994654  40.334809"
      ]
     },
     "execution_count": 113,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pc_mean_df = pd.DataFrame(\n",
    "    {'yes': pc_mean_yes, 'no':pc_mean_no}\n",
    ")\n",
    "pc_mean_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x7f3d6484ac18>"
      ]
     },
     "execution_count": 114,
     "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": [
    "pc_mean_df.reset_index(inplace=True)\n",
    "pc_mean_df['YEAR'] = pc_mean_df['YEAR'].astype(int)\n",
    "plt.plot('YEAR', 'yes', data=pc_mean_df, marker='', color='blue', linewidth=2)\n",
    "plt.plot('YEAR', 'no', data=pc_mean_df, marker='', color='olive', linewidth=2)\n",
    "plt.legend()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparing income bands and performance of students in public and private schools\n",
    "\n",
    "An interesting thing to do is to calculate how many students, within your monthly income band, are in public or private school, and how many of these are above the average of your own range.\n",
    "\n",
    "First we count how many students have each income band in each year:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "YEAR  FAMILY_INCOME \n",
       "1999  A_1_MW_or_less     1409\n",
       "      B_1_to_2_MW        6656\n",
       "      C_2_to_5_MW       21275\n",
       "      D_5_to_10_MW      21395\n",
       "      E_10_to_30_MW     10491\n",
       "Name: TOTAL_PUBLIC, dtype: int64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qty_per_year = enem.groupby(['YEAR', 'FAMILY_INCOME', 'SECONDARY_SCHOOL_2']).count()['MC_SCORE']\n",
    "qty_priv_per_year = qty_per_year.loc[qty_per_year.index.\\\n",
    "    get_level_values('SECONDARY_SCHOOL_2') == 'private'].xs('private', level=2).dropna().astype(int)\n",
    "qty_priv_per_year = qty_priv_per_year.loc[qty_priv_per_year.index.get_level_values('FAMILY_INCOME') != '*']\n",
    "qty_priv_per_year = qty_priv_per_year.rename('TOTAL_PRIVATE')\n",
    "\n",
    "qty_pub_per_year = qty_per_year.loc[qty_per_year.index.\\\n",
    "    get_level_values('SECONDARY_SCHOOL_2') == 'public'].xs('public', level=2).dropna().astype(int)\n",
    "qty_pub_per_year = qty_pub_per_year.loc[qty_pub_per_year.index.get_level_values('FAMILY_INCOME') != '*']\n",
    "qty_pub_per_year = qty_pub_per_year.rename('TOTAL_PUBLIC')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we take the average score for each income band per year:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "YEAR  FAMILY_INCOME \n",
       "1999  A_1_MW_or_less    33.968140\n",
       "      B_1_to_2_MW       36.554268\n",
       "      C_2_to_5_MW       41.410276\n",
       "      D_5_to_10_MW      47.226818\n",
       "      E_10_to_30_MW     53.802565\n",
       "Name: MEAN_GRADE, dtype: float64"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inc_mean_per_year = inc_mean_df.stack()\n",
    "inc_mean_per_year.index.names = ['YEAR', 'FAMILY_INCOME']\n",
    "inc_mean_per_year = inc_mean_per_year.rename(\"MEAN_GRADE\")\n",
    "inc_mean_per_year.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we count only the students whose score in the multiple choice test was higher than the average of the income band, taking into account whether the school is public or private:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "YEAR  FAMILY_INCOME \n",
       "1999  A_1_MW_or_less     537\n",
       "      B_1_to_2_MW       2537\n",
       "      C_2_to_5_MW       8527\n",
       "      D_5_to_10_MW      9096\n",
       "      E_10_to_30_MW     4589\n",
       "Name: COUNT_PUBLIC_ABOVE_MEAN, dtype: int64"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "enem_inc_public = enem[enem['SECONDARY_SCHOOL_2'] == 'public'][['YEAR', 'FAMILY_INCOME', 'MC_SCORE']]\n",
    "enem_inc_public = enem_inc_public.merge(inc_mean_per_year, how='left', on=['YEAR', 'FAMILY_INCOME'])\n",
    "qty_enem_inc_public = enem_inc_public.loc[enem_inc_public['MC_SCORE'] > enem_inc_public['MEAN_GRADE']].\\\n",
    "    groupby(['YEAR', 'FAMILY_INCOME']).count()['MC_SCORE']\n",
    "qty_enem_inc_public = qty_enem_inc_public.rename('COUNT_PUBLIC_ABOVE_MEAN')\n",
    "\n",
    "enem_inc_private = enem[enem['SECONDARY_SCHOOL_2'] == 'private'][['YEAR', 'FAMILY_INCOME', 'MC_SCORE']]\n",
    "enem_inc_private = enem_inc_private.merge(inc_mean_per_year, how='left', on=['YEAR', 'FAMILY_INCOME'])\n",
    "qty_enem_inc_private = enem_inc_private.loc[enem_inc_private['MC_SCORE'] > enem_inc_private['MEAN_GRADE']].\\\n",
    "    groupby(['YEAR', 'FAMILY_INCOME']).count()['MC_SCORE']\n",
    "qty_enem_inc_private = qty_enem_inc_private.rename('COUNT_PRIVATE_ABOVE_MEAN')\n",
    "\n",
    "qty_enem_inc_public.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we put them together and show the percentage:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>TOTAL_PRIVATE</th>\n",
       "      <th>COUNT_PRIVATE_ABOVE_MEAN</th>\n",
       "      <th>PER_PRIV_ABOVE_MEAN</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YEAR</th>\n",
       "      <th>FAMILY_INCOME</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">1999</th>\n",
       "      <th>A_1_MW_or_less</th>\n",
       "      <td>125</td>\n",
       "      <td>75</td>\n",
       "      <td>60.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>B_1_to_2_MW</th>\n",
       "      <td>542</td>\n",
       "      <td>302</td>\n",
       "      <td>55.719557</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_2_to_5_MW</th>\n",
       "      <td>3113</td>\n",
       "      <td>1801</td>\n",
       "      <td>57.854160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>D_5_to_10_MW</th>\n",
       "      <td>7284</td>\n",
       "      <td>4324</td>\n",
       "      <td>59.362987</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>E_10_to_30_MW</th>\n",
       "      <td>10617</td>\n",
       "      <td>6151</td>\n",
       "      <td>57.935387</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     TOTAL_PRIVATE  COUNT_PRIVATE_ABOVE_MEAN  \\\n",
       "YEAR FAMILY_INCOME                                             \n",
       "1999 A_1_MW_or_less            125                        75   \n",
       "     B_1_to_2_MW               542                       302   \n",
       "     C_2_to_5_MW              3113                      1801   \n",
       "     D_5_to_10_MW             7284                      4324   \n",
       "     E_10_to_30_MW           10617                      6151   \n",
       "\n",
       "                     PER_PRIV_ABOVE_MEAN  \n",
       "YEAR FAMILY_INCOME                        \n",
       "1999 A_1_MW_or_less            60.000000  \n",
       "     B_1_to_2_MW               55.719557  \n",
       "     C_2_to_5_MW               57.854160  \n",
       "     D_5_to_10_MW              59.362987  \n",
       "     E_10_to_30_MW             57.935387  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "priv_per_year = qty_priv_per_year.to_frame().\\\n",
    "    merge(qty_enem_inc_private, how='left', on=['YEAR', 'FAMILY_INCOME']).dropna()\n",
    "priv_per_year['COUNT_PRIVATE_ABOVE_MEAN'] = priv_per_year['COUNT_PRIVATE_ABOVE_MEAN'].astype(int)\n",
    "priv_per_year['PER_PRIV_ABOVE_MEAN'] = (priv_per_year['COUNT_PRIVATE_ABOVE_MEAN'] * 100)/ priv_per_year['TOTAL_PRIVATE']\n",
    "priv_per_year.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>TOTAL_PUBLIC</th>\n",
       "      <th>COUNT_PUBLIC_ABOVE_MEAN</th>\n",
       "      <th>PER_PUB_ABOVE_MEAN</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YEAR</th>\n",
       "      <th>FAMILY_INCOME</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"5\" valign=\"top\">1999</th>\n",
       "      <th>A_1_MW_or_less</th>\n",
       "      <td>1409</td>\n",
       "      <td>537</td>\n",
       "      <td>38.112136</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>B_1_to_2_MW</th>\n",
       "      <td>6656</td>\n",
       "      <td>2537</td>\n",
       "      <td>38.115986</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_2_to_5_MW</th>\n",
       "      <td>21275</td>\n",
       "      <td>8527</td>\n",
       "      <td>40.079906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>D_5_to_10_MW</th>\n",
       "      <td>21395</td>\n",
       "      <td>9096</td>\n",
       "      <td>42.514606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>E_10_to_30_MW</th>\n",
       "      <td>10491</td>\n",
       "      <td>4589</td>\n",
       "      <td>43.742255</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                     TOTAL_PUBLIC  COUNT_PUBLIC_ABOVE_MEAN  PER_PUB_ABOVE_MEAN\n",
       "YEAR FAMILY_INCOME                                                            \n",
       "1999 A_1_MW_or_less          1409                      537           38.112136\n",
       "     B_1_to_2_MW             6656                     2537           38.115986\n",
       "     C_2_to_5_MW            21275                     8527           40.079906\n",
       "     D_5_to_10_MW           21395                     9096           42.514606\n",
       "     E_10_to_30_MW          10491                     4589           43.742255"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pub_per_year = qty_pub_per_year.to_frame().\\\n",
    "    merge(qty_enem_inc_public, how='left', on=['YEAR', 'FAMILY_INCOME']).dropna()\n",
    "pub_per_year['COUNT_PUBLIC_ABOVE_MEAN'] = pub_per_year['COUNT_PUBLIC_ABOVE_MEAN'].astype(int)\n",
    "pub_per_year['PER_PUB_ABOVE_MEAN'] = (pub_per_year['COUNT_PUBLIC_ABOVE_MEAN'] * 100)/ pub_per_year['TOTAL_PUBLIC']\n",
    "pub_per_year.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's now take each of the income bands and show the proportion of candidates each year by school type:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "mean_per_year = pd.DataFrame()\n",
    "mean_per_year['PER_PUB_ABOVE_MEAN'] = pub_per_year['PER_PUB_ABOVE_MEAN']\n",
    "mean_per_year['PER_PRIV_ABOVE_MEAN'] = priv_per_year['PER_PRIV_ABOVE_MEAN']\n",
    "\n",
    "df_A = mean_per_year.loc[mean_per_year.index.get_level_values('FAMILY_INCOME') == 'A_1_MW_or_less'].xs('A_1_MW_or_less', level=1)\n",
    "df_B = mean_per_year.loc[mean_per_year.index.get_level_values('FAMILY_INCOME') == 'B_1_to_2_MW'].xs('B_1_to_2_MW', level=1)\n",
    "df_C = mean_per_year.loc[mean_per_year.index.get_level_values('FAMILY_INCOME') == 'C_2_to_5_MW'].xs('C_2_to_5_MW', level=1)\n",
    "df_D = mean_per_year.loc[mean_per_year.index.get_level_values('FAMILY_INCOME') == 'D_5_to_10_MW'].xs('D_5_to_10_MW', level=1)\n",
    "df_E = mean_per_year.loc[mean_per_year.index.get_level_values('FAMILY_INCOME') == 'E_10_to_30_MW'].xs('E_10_to_30_MW', level=1)\n",
    "df_F = mean_per_year.loc[mean_per_year.index.get_level_values('FAMILY_INCOME') == 'F_30_to_50_MW'].xs('F_30_to_50_MW', level=1)\n",
    "df_G = mean_per_year.loc[mean_per_year.index.get_level_values('FAMILY_INCOME') == 'G_50+_MW'].xs('G_50+_MW', level=1)\n",
    "df_H = mean_per_year.loc[mean_per_year.index.get_level_values('FAMILY_INCOME') == 'H_no_income'].xs('H_no_income', level=1)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3d4d219940>"
      ]
     },
     "execution_count": 28,
     "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": [
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "df_A.plot(kind='bar', title=\"1 Minimum Wage or less per month\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3d4d3a62b0>"
      ]
     },
     "execution_count": 29,
     "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": [
    "df_B.plot(kind='bar', title=\"From 1 to 2 Minimum Wages per month\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3d65ff1080>"
      ]
     },
     "execution_count": 30,
     "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": [
    "df_C.plot(kind='bar', title=\"From 2 to 5 Minimum Wages per month\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3d65f8a358>"
      ]
     },
     "execution_count": 31,
     "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": [
    "df_D.plot(kind='bar', title=\"From 5 to 10 Minimum Wages per month\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3d65e9db00>"
      ]
     },
     "execution_count": 32,
     "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": [
    "df_E.plot(kind='bar', title=\"From 10 to 30 Minimum Wages per month\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3d65e239e8>"
      ]
     },
     "execution_count": 33,
     "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": [
    "df_F.plot(kind='bar', title=\"From 30 to 50 Minimum Wages per month\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3d65daad30>"
      ]
     },
     "execution_count": 34,
     "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": [
    "df_G.plot(kind='bar', title=\"More than 50 Minimum Wages per month\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It is interesting to note that, regardless of the student's income band, proportionately more students from private schools had an objective test score above average."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Comparing parental schooling and performance of public and private school students\n",
    "\n",
    "Another interesting thing to do is to use the same methodology to compare the average score of students whose parents have higher education and whose both parents do not have and compare their respective performances above or below average with the fact that they studied in public schools or private.\n",
    "\n",
    "First we take the average of each year and we group with respect to the two parents have higher education and the two do not have:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "YEAR  PARENTS_SCHOOLING\n",
       "1999  both                 60.438033\n",
       "      none                 45.683902\n",
       "2000  both                 62.987877\n",
       "      none                 49.936372\n",
       "2001  both                 53.361863\n",
       "Name: MEAN_GRADE, dtype: float64"
      ]
     },
     "execution_count": 82,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "mask_parents_education = ((enem['FATHER_SCHOOLING'] == 'G-complete_higher_education') | \\\n",
    "      (enem['FATHER_SCHOOLING'] == 'H-postgraduate')) & \\\n",
    "     ((enem['MOTHER_SCHOOLING'] == 'G-complete_higher_education') | \\\n",
    "      (enem['MOTHER_SCHOOLING'] == 'H-postgraduate'))\n",
    "\n",
    "mean_education_both = enem[['YEAR', 'MC_SCORE']].loc[mask_parents_education]\n",
    "mean_education_both = mean_education_both.groupby('YEAR').mean()['MC_SCORE']\n",
    "mean_education_both = mean_education_both.rename(\"MEAN_GRADE\").to_frame()\n",
    "mean_education_both['PARENTS_SCHOOLING'] = 'both'\n",
    "\n",
    "mean_education_none = enem[['YEAR', 'MC_SCORE']].loc[~mask_parents_education]\n",
    "mean_education_none = mean_education_none.groupby('YEAR').mean()['MC_SCORE']\n",
    "mean_education_none = mean_education_none.rename(\"MEAN_GRADE\").to_frame()\n",
    "mean_education_none['PARENTS_SCHOOLING'] = 'none'\n",
    "\n",
    "mean_education_ps = pd.concat([mean_education_both, mean_education_none])\n",
    "mean_education_ps.reset_index(inplace=True)\n",
    "mean_education_ps.sort_values(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "mean_education_ps.set_index(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "mean_education_ps = mean_education_ps.squeeze().rename(\"MEAN_GRADE\")\n",
    "\n",
    "mean_education_ps.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we count how many students in each of these bands did public or private school:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "YEAR  PARENTS_SCHOOLING\n",
       "1999  both                  2004\n",
       "      none                 60892\n",
       "2000  both                   950\n",
       "      none                 24338\n",
       "2001  both                   798\n",
       "Name: TOTAL_PUBLIC, dtype: int64"
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "qty_both_public_per_year = enem[['YEAR', 'MC_SCORE', 'SECONDARY_SCHOOL_2']].\\\n",
    "    loc[mask_parents_education & (enem['SECONDARY_SCHOOL_2'] == 'public')].\\\n",
    "    groupby('YEAR').count()['MC_SCORE']\n",
    "qty_both_public_per_year = qty_both_public_per_year.to_frame()\n",
    "qty_both_public_per_year['PARENTS_SCHOOLING'] = 'both'\n",
    "qty_none_public_per_year = enem[['YEAR', 'MC_SCORE', 'SECONDARY_SCHOOL_2']].\\\n",
    "    loc[~mask_parents_education & (enem['SECONDARY_SCHOOL_2'] == 'public')]\\\n",
    "    .groupby('YEAR').count()['MC_SCORE']\n",
    "qty_none_public_per_year = qty_none_public_per_year.to_frame()\n",
    "qty_none_public_per_year['PARENTS_SCHOOLING'] = 'none'\n",
    "qty_public_per_year_ps = pd.concat([qty_both_public_per_year, qty_none_public_per_year])\n",
    "qty_public_per_year_ps.reset_index(inplace=True)\n",
    "qty_public_per_year_ps.sort_values(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "qty_public_per_year_ps.set_index(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "qty_public_per_year_ps = qty_public_per_year_ps.squeeze().rename(\"TOTAL_PUBLIC\")\n",
    "\n",
    "\n",
    "qty_both_private_per_year = enem[['YEAR', 'MC_SCORE', 'SECONDARY_SCHOOL_2']].\\\n",
    "    loc[mask_parents_education & (enem['SECONDARY_SCHOOL_2'] == 'private')].\\\n",
    "    groupby('YEAR').count()['MC_SCORE']\n",
    "qty_both_private_per_year = qty_both_private_per_year.to_frame()\n",
    "qty_both_private_per_year['PARENTS_SCHOOLING'] = 'both'\n",
    "qty_none_private_per_year = enem[['YEAR', 'MC_SCORE', 'SECONDARY_SCHOOL_2']].\\\n",
    "    loc[~mask_parents_education & (enem['SECONDARY_SCHOOL_2'] == 'private')]\\\n",
    "    .groupby('YEAR').count()['MC_SCORE']\n",
    "qty_none_private_per_year = qty_none_private_per_year.to_frame()\n",
    "qty_none_private_per_year['PARENTS_SCHOOLING'] = 'none'\n",
    "qty_private_per_year_ps = pd.concat([qty_both_private_per_year, qty_none_private_per_year])\n",
    "qty_private_per_year_ps.reset_index(inplace=True)\n",
    "qty_private_per_year_ps.sort_values(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "qty_private_per_year_ps.set_index(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "qty_private_per_year_ps = qty_private_per_year_ps.squeeze().rename(\"TOTAL_PRIVATE\")\n",
    "\n",
    "\n",
    "qty_public_per_year_ps.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then we count how many students in each of these bands, taking into account the public or private school, had above average scores:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [],
   "source": [
    "enem_both_public_ps = enem[['YEAR', 'MC_SCORE']].loc[mask_parents_education & (enem['SECONDARY_SCHOOL_2'] == 'public')] \n",
    "enem_both_public_ps = enem_both_public_ps.merge(mean_education_both, how='left', on=['YEAR'])\n",
    "enem_both_public_ps = enem_both_public_ps.loc[enem_both_public_ps['MC_SCORE'] > enem_both_public_ps['MEAN_GRADE']].\\\n",
    "    groupby(['YEAR']).count()['MC_SCORE']\n",
    "enem_both_public_ps = enem_both_public_ps.rename('COUNT_ABOVE')\n",
    "enem_both_public_ps = enem_both_public_ps.to_frame()\n",
    "enem_both_public_ps['PARENTS_SCHOOLING'] = 'both'\n",
    "enem_none_public_ps = enem[['YEAR', 'MC_SCORE']].loc[~mask_parents_education & (enem['SECONDARY_SCHOOL_2'] == 'public')] \n",
    "enem_none_public_ps = enem_none_public_ps.merge(mean_education_none, how='left', on=['YEAR'])\n",
    "enem_none_public_ps = enem_none_public_ps.loc[enem_none_public_ps['MC_SCORE'] > enem_none_public_ps['MEAN_GRADE']].\\\n",
    "    groupby(['YEAR']).count()['MC_SCORE']\n",
    "enem_none_public_ps = enem_none_public_ps.rename('COUNT_ABOVE')\n",
    "enem_none_public_ps = enem_none_public_ps.to_frame()\n",
    "enem_none_public_ps['PARENTS_SCHOOLING'] = 'none'\n",
    "above_mean_public_per_year_ps = pd.concat([enem_both_public_ps, enem_none_public_ps])\n",
    "above_mean_public_per_year_ps.reset_index(inplace=True)\n",
    "above_mean_public_per_year_ps.sort_values(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "above_mean_public_per_year_ps.set_index(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "above_mean_public_per_year_ps = above_mean_public_per_year_ps.squeeze().rename(\"COUNT_PUBLIC_ABOVE_MEAN\")\n",
    "\n",
    "enem_both_private_ps = enem[['YEAR', 'MC_SCORE']].loc[mask_parents_education & (enem['SECONDARY_SCHOOL_2'] == 'private')] \n",
    "enem_both_private_ps = enem_both_private_ps.merge(mean_education_both, how='left', on=['YEAR'])\n",
    "enem_both_private_ps = enem_both_private_ps.loc[enem_both_private_ps['MC_SCORE'] > enem_both_private_ps['MEAN_GRADE']].\\\n",
    "    groupby(['YEAR']).count()['MC_SCORE']\n",
    "enem_both_private_ps = enem_both_private_ps.rename('COUNT_ABOVE')\n",
    "enem_both_private_ps = enem_both_private_ps.to_frame()\n",
    "enem_both_private_ps['PARENTS_SCHOOLING'] = 'both'\n",
    "enem_none_private_ps = enem[['YEAR', 'MC_SCORE']].loc[~mask_parents_education & (enem['SECONDARY_SCHOOL_2'] == 'private')] \n",
    "enem_none_private_ps = enem_none_private_ps.merge(mean_education_none, how='left', on=['YEAR'])\n",
    "enem_none_private_ps = enem_none_private_ps.loc[enem_none_private_ps['MC_SCORE'] > enem_none_private_ps['MEAN_GRADE']].\\\n",
    "    groupby(['YEAR']).count()['MC_SCORE']\n",
    "enem_none_private_ps = enem_none_private_ps.rename('COUNT_ABOVE')\n",
    "enem_none_private_ps = enem_none_private_ps.to_frame()\n",
    "enem_none_private_ps['PARENTS_SCHOOLING'] = 'none'\n",
    "above_mean_private_per_year_ps = pd.concat([enem_both_private_ps, enem_none_private_ps])\n",
    "above_mean_private_per_year_ps.reset_index(inplace=True)\n",
    "above_mean_private_per_year_ps.sort_values(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "above_mean_private_per_year_ps.set_index(['YEAR', 'PARENTS_SCHOOLING'], inplace=True)\n",
    "above_mean_private_per_year_ps = above_mean_private_per_year_ps.squeeze().rename(\"COUNT_PRIVATE_ABOVE_MEAN\")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we put it all together to get the percentage:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "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>TOTAL_PUBLIC</th>\n",
       "      <th>COUNT_PUBLIC_ABOVE_MEAN</th>\n",
       "      <th>PER_PUB_ABOVE_MEAN</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>YEAR</th>\n",
       "      <th>PARENTS_SCHOOLING</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">1999</th>\n",
       "      <th>both</th>\n",
       "      <td>2004</td>\n",
       "      <td>918</td>\n",
       "      <td>45.808383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>none</th>\n",
       "      <td>60892</td>\n",
       "      <td>24037</td>\n",
       "      <td>39.474808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th rowspan=\"2\" valign=\"top\">2000</th>\n",
       "      <th>both</th>\n",
       "      <td>950</td>\n",
       "      <td>526</td>\n",
       "      <td>55.368421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>none</th>\n",
       "      <td>24338</td>\n",
       "      <td>10678</td>\n",
       "      <td>43.873778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2001</th>\n",
       "      <th>both</th>\n",
       "      <td>798</td>\n",
       "      <td>283</td>\n",
       "      <td>35.463659</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                        TOTAL_PUBLIC  COUNT_PUBLIC_ABOVE_MEAN  \\\n",
       "YEAR PARENTS_SCHOOLING                                          \n",
       "1999 both                       2004                      918   \n",
       "     none                      60892                    24037   \n",
       "2000 both                        950                      526   \n",
       "     none                      24338                    10678   \n",
       "2001 both                        798                      283   \n",
       "\n",
       "                        PER_PUB_ABOVE_MEAN  \n",
       "YEAR PARENTS_SCHOOLING                      \n",
       "1999 both                        45.808383  \n",
       "     none                        39.474808  \n",
       "2000 both                        55.368421  \n",
       "     none                        43.873778  \n",
       "2001 both                        35.463659  "
      ]
     },
     "execution_count": 98,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pub_per_year = qty_public_per_year_ps.to_frame().\\\n",
    "    merge(above_mean_public_per_year_ps, how='left', on=['YEAR', 'PARENTS_SCHOOLING']).dropna()\n",
    "pub_per_year['PER_PUB_ABOVE_MEAN'] = (pub_per_year['COUNT_PUBLIC_ABOVE_MEAN'] * 100)/ pub_per_year['TOTAL_PUBLIC']\n",
    "\n",
    "\n",
    "priv_per_year = qty_private_per_year_ps.to_frame().\\\n",
    "    merge(above_mean_private_per_year_ps, how='left', on=['YEAR', 'PARENTS_SCHOOLING']).dropna()\n",
    "priv_per_year['PER_PRIV_ABOVE_MEAN'] = (priv_per_year['COUNT_PRIVATE_ABOVE_MEAN'] * 100)/ priv_per_year['TOTAL_PRIVATE']\n",
    "pub_per_year.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "And finally we put it all together in a DataFrame and we plot the percentage of private and public school students who had above average scores taking into account the parents' schooling situation:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f3d64cc11d0>"
      ]
     },
     "execution_count": 110,
     "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": [
    "mean_per_year = pd.DataFrame()\n",
    "mean_per_year['PER_PUB_BOTH_ABOVE_MEAN'] = pub_per_year['PER_PUB_ABOVE_MEAN'].\\\n",
    "    loc[pub_per_year.index.get_level_values('PARENTS_SCHOOLING') == 'both'].xs('both', level=1)\n",
    "mean_per_year['PER_PRIV_BOTH_ABOVE_MEAN'] = priv_per_year['PER_PRIV_ABOVE_MEAN'].\\\n",
    "    loc[priv_per_year.index.get_level_values('PARENTS_SCHOOLING') == 'both'].xs('both', level=1)\n",
    "mean_per_year['PER_PUB_NONE_ABOVE_MEAN'] = pub_per_year['PER_PUB_ABOVE_MEAN'].\\\n",
    "    loc[pub_per_year.index.get_level_values('PARENTS_SCHOOLING') == 'none'].xs('none', level=1)\n",
    "mean_per_year['PER_PRIV_NONE_ABOVE_MEAN'] = priv_per_year['PER_PRIV_ABOVE_MEAN'].\\\n",
    "    loc[priv_per_year.index.get_level_values('PARENTS_SCHOOLING') == 'none'].xs('none', level=1)\n",
    "\n",
    "mean_per_year.\\\n",
    "    plot(title=\"Percentage of candidates performing above mean\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Conclusion and further proposals\n",
    "\n",
    "The data presented were very useful to show correlations between the notes and the socio-economic situation of the candidates. The average grade of students was expected to grow along with the monthly family income range, as well as we expected that the scores of those who studied in private high school were higher than the grade of those who studied in public schools, as well as we expected a correlation between a better note and water and light infrastructure. What we did not expect is such a low proportion of public school students whose parents have tertiary education who scored above average.\n",
    "\n",
    "We also conclude that the Pandas tool is very useful for making data inferences. The difficult part was to consolidate and normalize the data so that the data of every year could have the same treatment.\n",
    "\n",
    "Other studies that we suggest for those who are interested:\n",
    "\n",
    "* Study data after 2008\n",
    "\n",
    "* Look for other correlations, such as readings and editorial note\n",
    "\n",
    "* Expand metadata table to load more data\n",
    "\n",
    "Any contribution to this study is welcome! You can make a push request and submit changes at any time to the https://github.com/lucasbracher/enem repository.\n"
   ]
  }
 ],
 "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": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}