{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Notebook réalisé par [Pierre-Alexandre Aranega](mailto:pierre-alexandre.aranega@ensae.fr) et [Charles Cros](mailto:charles.cros@ensae.fr) dans le cadre du projet **Python pour l'économiste** (ENSAE 2A - 2019-2020)."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Introduction\n",
"\n",
"Le 3 décembre 2019, l'OCDE a publié les résultats de l'enquête PISA (*Programme for International Student Assessment*) menée en 2018 dans 79 pays, sur des élèves de 15 ans, et répétée tous les trois ans depuis l'an 2000. Ces résultats sont très commentés et sont considérés comme un indicateur pertinent de l'évolution du niveau scolaire des élèves, de manière absolue et de manière relative (en comparaison avec les pays voisins). Quelques articles sur la question :\n",
"\n",
"* [Nouveau classement PISA : l'école française, cet élève moyen qui ne progresse pas](http://www.leparisien.fr/societe/nouveau-classement-pisa-l-ecole-francaise-cet-eleve-moyen-qui-ne-progresse-pas-03-12-2019-8208645.php) (Le Parisien, 3 déc. 2019)\n",
"* [Enquête PISA : dix résultats pour situer les élèves français](https://www.lesechos.fr/politique-societe/societe/enquete-pisa-dix-resultats-pour-situer-les-eleves-francais-1153200) (Les Échos, 3 déc. 2019)\n",
"* [La France aura-t-elle un jour son choc Pisa ?](https://www.lepoint.fr/education/systeme-educatif-la-france-aura-t-elle-un-jour-son-choc-pisa-03-12-2019-2351038_3584.php) (Le Point, 3 déc. 2019)\n",
"* [Enquête PISA 2018 : stabilité des résultats des élèves français de 15 ans](https://www.education.gouv.fr/cid147361/enquete-pisa-2018-stabilite-des-resultats-des-eleves-francais-de-15-ans.html) (Communiqué de presse, Jean-Michel Blanquer, 3 déc. 2019)\n",
"* [PISA 2018 : stabilité des résultats en compréhension de l'écrit](https://www.education.gouv.fr/cid54176/pisa-2018-stabilite-des-resultats-en-comprehension-de-l-ecrit.html) (Note d'information, Ministère de l'Éducation, 3 déc. 2019)\n",
"\n",
"Suivant les articles ou les publications officielles, il est difficile de savoir s'il faut ou non se réjouir de ces résultats. Les commentaires ne sont jamais neutres : certains s'inquiètent d'une nouvelle baisse des résultats français ; d'autres relativisent, observent que cette baisse est tout de même moins forte que la précédente, et que la France reste au-dessus de la moyenne des pays de l'OCDE...\n",
"\n",
"L'objectif de ce projet est de :\n",
"* Décrire les **résultats de l'édition 2018** de cette enquête, en portant une attention particulière au cas de la France.\n",
"* Décrire l'**évolution des résultats** depuis 2000,\n",
"* Explorer les **corrélations entre les variables**, et **avec d'autres variables**, notamment macroéconomiques.\n",
"\n",
"Pour information, toutes les publications de l'OCDE sont disponibles [à cette adresse](https://www.oecd.org/pisa/publications/). L'enquête PISA couvre de très nombreux sujets, divers et passionnants, qui ne se limitent pas à la seule question de la performance. Soucieux de rester concis, on ne pourra évidemment pas tous les aborder ici.\n",
"\n",
"## Table des matières"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
run previous cell, wait for 2 seconds
\n",
""
],
"text/plain": [
""
]
},
"execution_count": 1,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"from jyquickhelper import add_notebook_menu\n",
"add_notebook_menu()"
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {},
"outputs": [],
"source": [
"import numpy as np\n",
"import pandas as pd\n",
"import urllib\n",
"import bs4\n",
"import matplotlib.pyplot as plt\n",
"import seaborn as sns\n",
"import plotly.express as px\n",
"import plotly.graph_objects as go\n",
"import statsmodels.api as sm\n",
"from sklearn.cluster import KMeans"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## I. Import de données par scrapping"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### A. Données PISA (source : Wikipedia)\n",
"Les données sont issues de la page Wikipedia [Programme for International Student Assessment](https://en.wikipedia.org/wiki/Programme_for_International_Student_Assessment)."
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {},
"outputs": [],
"source": [
"url_wikipedia = \"https://en.wikipedia.org/wiki/Programme_for_International_Student_Assessment\"\n",
"request_text_PISA = urllib.request.urlopen(url_wikipedia).read()\n",
"page_PISA = bs4.BeautifulSoup(request_text_PISA, \"lxml\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 1. Année 2018 (résultats publiés en décembre 2019)\n",
"Cette section se concentre sur le paragraphe [PISA 2018 ranking summary](https://en.wikipedia.org/wiki/Programme_for_International_Student_Assessment#PISA_2018_ranking_summary) qui donne les résultats de l'année 2018 dans un format particulier.\n",
"\n",
"On aurait aimé disposer d'un id, d'une classe ou d'un titre permettant de cibler facilement le bon tableau. Faute de mieux, on se résout à utiliser une spécificité qui est la largeur de table, fixée à 240px pour les sous-tables qui nous intéressent ici."
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Country
\n",
"
Score
\n",
"
Year
\n",
"
Subject
\n",
"
\n",
" \n",
" \n",
"
\n",
"
24
\n",
"
France
\n",
"
495
\n",
"
2018
\n",
"
Maths
\n",
"
\n",
"
\n",
"
23
\n",
"
France
\n",
"
493
\n",
"
2018
\n",
"
Science
\n",
"
\n",
"
\n",
"
22
\n",
"
France
\n",
"
493
\n",
"
2018
\n",
"
Reading
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Country Score Year Subject\n",
"24 France 495 2018 Maths\n",
"23 France 493 2018 Science\n",
"22 France 493 2018 Reading"
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# On se concentre sur les tables ayant la particularité de faire 240px de largeur...\n",
"tables_2018 = page_PISA.findAll(\"table\", width=\"240\")\n",
"tables_2018 = pd.read_html(str(tables_2018))\n",
"\n",
"# On définit les headers manuellement afin d'harmoniser avec les intitulés de la partie suivante.\n",
"liste_matieres = [\"Maths\", \"Science\", \"Reading\"] \n",
"\n",
"# On fusionne ces tables dans un dataframe\n",
"PISA_2018 = pd.DataFrame()\n",
"for id_table in range(0,3):\n",
" table = tables_2018[id_table].iloc[:,1:3]\n",
" table.columns = [\"Country\", \"Score\"]\n",
" table[\"Year\"] = 2018\n",
" table[\"Subject\"] = liste_matieres[id_table]\n",
" \n",
" PISA_2018 = pd.concat([PISA_2018,table])\n",
" \n",
"# Les résultats de la France en 2018\n",
"PISA_2018[PISA_2018[\"Country\"]==\"France\"] "
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 2. Années 2000, 2003, 2006, 2009, 2012, 2015\n",
"\n",
"Cette section se concentre sur le paragraphe [Rankings comparison 2003 - 2015](https://en.wikipedia.org/wiki/Programme_for_International_Student_Assessment#Rankings_comparison_2003_-_2015) qui regroupe les résultats des années précédentes.\n",
"\n",
"De même que ci-dessus, nous n'avons pas d'id unique pour se référer aux tableaux qui nous intéressent. On choisit de ne garder que les tables ayant au moins 9 colonnes, ce qui permet de filtrer les trois tableaux qui nous préoccupent de manière un peu brutale, mais rapide."
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Country
\n",
"
Score
\n",
"
Subject
\n",
"
Year
\n",
"
\n",
" \n",
" \n",
"
\n",
"
24
\n",
"
France
\n",
"
495
\n",
"
Maths
\n",
"
2018
\n",
"
\n",
"
\n",
"
23
\n",
"
France
\n",
"
493
\n",
"
Science
\n",
"
2018
\n",
"
\n",
"
\n",
"
22
\n",
"
France
\n",
"
493
\n",
"
Reading
\n",
"
2018
\n",
"
\n",
"
\n",
"
22
\n",
"
France
\n",
"
493
\n",
"
Maths
\n",
"
2015
\n",
"
\n",
"
\n",
"
95
\n",
"
France
\n",
"
495
\n",
"
Maths
\n",
"
2012
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Country Score Subject Year\n",
"24 France 495 Maths 2018\n",
"23 France 493 Science 2018\n",
"22 France 493 Reading 2018\n",
"22 France 493 Maths 2015\n",
"95 France 495 Maths 2012"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"all_tables = page_PISA.findAll(\"table\")\n",
"all_tables = pd.read_html(str(all_tables), header=[0,1], na_values = [\"—\", \"‡\"])\n",
"\n",
"# On ne s'intéresse qu'aux trois tables ayant + de 9 colonnes.\n",
"\n",
"PISA = PISA_2018\n",
"for table in all_tables:\n",
" if len(table.columns) > 8:\n",
" subject = table.columns[0][0]\n",
" table.columns = table.columns.droplevel(0)\n",
" # On supprime toutes les colonnes de \"rang\" qui nous sont inutiles.\n",
" for column in table.columns:\n",
" if table[column][0] == \"Rank\":\n",
" table = table.drop(columns=column)\n",
" # On supprime les 2 premières lignes : \"moyenne OCDE\" et sous-titre \"Score\" qui nous sont inutiles.\n",
" table = table.drop([0,1])\n",
" # On dé-pivote les années.\n",
" table = table.melt(id_vars=\"Country\",var_name=\"Year\",value_name=\"Score\")\n",
" # On ajoute la matière en colonne.\n",
" table[\"Subject\"] = subject\n",
" # On fusionne avec le tableau complet, qui avait commencé à être créé pour l'année 2018.\n",
" PISA = pd.concat([PISA,table], sort=True) \n",
"\n",
"PISA[PISA[\"Country\"]==\"France\"].head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 3. Nettoyage de la base de données produite"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"# Réinitialisation de l'index du dataframe, suite aux diverses concaténations / suppressions.\n",
"PISA = PISA.reset_index(drop=True)\n",
"\n",
"# Suppression des scores N/A\n",
"PISA = PISA[pd.notnull(PISA[\"Score\"])]\n",
"\n",
"# Conversion de certains formats puis tri.\n",
"PISA[\"Year\"]=PISA[\"Year\"].astype(int)\n",
"PISA[\"Score\"]=PISA[\"Score\"].astype(int)\n",
"PISA = PISA.sort_values(by=[\"Year\", \"Subject\", \"Score\"], ascending=False)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### B. Données complémentaires\n",
"On commence avant tout par régler quelques cas particuliers qui pourraient poser problème lors des jointures à venir."
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"# On ignore le cas particulier de Buenos Aires, traitée séparément du reste de l'Argentine.\n",
"PISA = PISA[PISA[\"Country\"]!=\"Argentina CABA[b]\"]\n",
"\n",
"# On harmonise le nom de certains pays qui varient selon les bases.\n",
"PISA.loc[PISA[\"Country\"].str.contains(\"China\"), \"Country\"] = \"China\"\n",
"PISA[\"Country\"] = PISA[\"Country\"].replace(\"Hong Kong, China\", \"Hong Kong\")\n",
"PISA[\"Country\"] = PISA[\"Country\"].replace(\"Czech Republic\", \"Czechia\")\n",
"PISA[\"Country\"] = PISA[\"Country\"].replace([\"South Korea\", \"Korea\"], \"Korea, South\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 1. Pays membres de l'OCDE (source : Wikipedia)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On souhaite disposer, dans la suite de ce devoir, de la [liste des pays membres de l'OCDE](https://en.wikipedia.org/wiki/OECD#Current_members). Ils constitueront notre pool de comparaison au fil du temps."
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {},
"outputs": [],
"source": [
"url_wikipedia_OECD = \"https://en.wikipedia.org/wiki/OECD\"\n",
"request_text_OECD = urllib.request.urlopen(url_wikipedia_OECD).read()\n",
"page_OECD = bs4.BeautifulSoup(request_text_OECD, \"lxml\")"
]
},
{
"cell_type": "code",
"execution_count": 9,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Les 36 pays membres de l'OCDE sont : Australia, Austria, Belgium, Canada, Chile, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, South Korea, Latvia, Lithuania, Luxembourg, Mexico, Netherlands, New Zealand, Norway, Poland, Portugal, Slovakia, Slovenia, Spain, Sweden, Switzerland, Turkey, United Kingdom, United States\n"
]
}
],
"source": [
"OECD_countries = page_OECD.findAll(\"table\", class_=\"wikitable sortable\")\n",
"OECD_countries = pd.read_html(str(OECD_countries), header=0)[0][\"Country\"]\n",
"print(\"Les\", len(OECD_countries), \"pays membres de l'OCDE sont :\", OECD_countries.str.cat(sep=\", \"))"
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Country
\n",
"
Score
\n",
"
Subject
\n",
"
Year
\n",
"
is_OECD
\n",
"
\n",
" \n",
" \n",
"
\n",
"
78
\n",
"
China
\n",
"
590
\n",
"
Science
\n",
"
2018
\n",
"
0
\n",
"
\n",
"
\n",
"
79
\n",
"
Singapore
\n",
"
551
\n",
"
Science
\n",
"
2018
\n",
"
0
\n",
"
\n",
"
\n",
"
80
\n",
"
Macau
\n",
"
544
\n",
"
Science
\n",
"
2018
\n",
"
0
\n",
"
\n",
"
\n",
"
81
\n",
"
Estonia
\n",
"
530
\n",
"
Science
\n",
"
2018
\n",
"
1
\n",
"
\n",
"
\n",
"
82
\n",
"
Japan
\n",
"
529
\n",
"
Science
\n",
"
2018
\n",
"
1
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Country Score Subject Year is_OECD\n",
"78 China 590 Science 2018 0\n",
"79 Singapore 551 Science 2018 0\n",
"80 Macau 544 Science 2018 0\n",
"81 Estonia 530 Science 2018 1\n",
"82 Japan 529 Science 2018 1"
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Ajoutons cette information à la base principale\n",
"PISA[\"is_OECD\"] = PISA.apply(lambda row : (row[\"Country\"] in OECD_countries.values)*1, axis = 1 )\n",
"PISA.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 2. Codes iso3 des pays participants (source : CIA)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On aura également besoin d'accéder au code ISO-3 (identifiant à trois lettres) de chaque pays, notamment pour l'utilisation du package servant à tracer une carte choroplèthe. Utilisons, pour varier, le site de la [CIA](https://www.cia.gov/library/publications/the-world-factbook/appendix/appendix-d.html). Le scrapping s'avère facilité : il n'y a qu'une seule table sur toute la page."
]
},
{
"cell_type": "code",
"execution_count": 11,
"metadata": {},
"outputs": [],
"source": [
"url_CIA_iso3codes = \"https://www.cia.gov/library/publications/the-world-factbook/appendix/appendix-d.html\"\n",
"request_text_CIA = urllib.request.urlopen(url_CIA_iso3codes).read()\n",
"page_iso3codes = bs4.BeautifulSoup(request_text_CIA, \"lxml\")"
]
},
{
"cell_type": "code",
"execution_count": 12,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"250 codes ISO3 ont bien été importés.\n"
]
}
],
"source": [
"table_iso3codes = page_iso3codes.findAll(\"table\")\n",
"table_iso3codes = pd.read_html(str(table_iso3codes), header=0, na_values=\"-\")[0]\n",
"table_iso3codes = table_iso3codes[[\"entity\", \"iso 3166.1\"]]\n",
"table_iso3codes.columns = [\"Country\", \"iso3code\"]\n",
"print(str(len(table_iso3codes[\"iso3code\"].unique())) + \" codes ISO3 ont bien été importés.\")"
]
},
{
"cell_type": "code",
"execution_count": 13,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Country
\n",
"
Score
\n",
"
Subject
\n",
"
Year
\n",
"
is_OECD
\n",
"
iso3code
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
China
\n",
"
590
\n",
"
Science
\n",
"
2018
\n",
"
0
\n",
"
CHN
\n",
"
\n",
"
\n",
"
1
\n",
"
Singapore
\n",
"
551
\n",
"
Science
\n",
"
2018
\n",
"
0
\n",
"
SGP
\n",
"
\n",
"
\n",
"
2
\n",
"
Macau
\n",
"
544
\n",
"
Science
\n",
"
2018
\n",
"
0
\n",
"
MAC
\n",
"
\n",
"
\n",
"
3
\n",
"
Estonia
\n",
"
530
\n",
"
Science
\n",
"
2018
\n",
"
1
\n",
"
EST
\n",
"
\n",
"
\n",
"
4
\n",
"
Japan
\n",
"
529
\n",
"
Science
\n",
"
2018
\n",
"
1
\n",
"
JPN
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Country Score Subject Year is_OECD iso3code\n",
"0 China 590 Science 2018 0 CHN\n",
"1 Singapore 551 Science 2018 0 SGP\n",
"2 Macau 544 Science 2018 0 MAC\n",
"3 Estonia 530 Science 2018 1 EST\n",
"4 Japan 529 Science 2018 1 JPN"
]
},
"execution_count": 13,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Ajoutons cette information à la base principale.\n",
"PISA = PISA.merge(table_iso3codes, on=\"Country\", how=\"left\")\n",
"PISA.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 3. Fonction d'import depuis le site de la Banque Mondiale"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Étant donné que l'on va importer successivement plusieurs indicateurs depuis le site de la Banque Mondiale, on définit une fonction pour cela. L'organisme a le bon goût de proposer un API d'export au format Excel de tous ses indicateurs, avec toujours le même format."
]
},
{
"cell_type": "code",
"execution_count": 14,
"metadata": {},
"outputs": [],
"source": [
"def import_from_World_Bank(ref_indicator, name):\n",
"\n",
" # Lecture depuis le site Banque Mondiale\n",
" url_WB = \"http://api.worldbank.org/v2/en/indicator/\"+ ref_indicator +\"?downloadformat=excel\"\n",
" table = pd.read_excel(url_WB, header=3)\n",
"\n",
" # Remplissage des N/A avec la dernière valeur disponible.\n",
" sub_table = table[table.columns[4:]]\n",
" sub_table = sub_table.fillna(method=\"ffill\", axis=\"columns\")\n",
" sub_table = sub_table.fillna(method=\"bfill\", axis=\"columns\")\n",
" table[table.columns[4:]] = sub_table\n",
"\n",
" # Dé-pivotage des colonnes d'année\n",
" table = table.melt(id_vars=table.columns[0:4], var_name=\"Year\", value_name=name)\n",
"\n",
" # Filtrage des colonnes qui nous intéressent (pays, année, valeur) puis renommage\n",
" table = table[table.columns[[1,4,5]]]\n",
" table.columns = [\"iso3code\", \"Year\", name]\n",
"\n",
" # Changement de format puis tri par année\n",
" table[\"Year\"] = table[\"Year\"].astype(int)\n",
" table = table.sort_values(by=\"Year\", ascending=False)\n",
" \n",
" return table"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 4. PIB par tête, par année et par pays (source : Banque Mondiale)\n",
"Source : [GDP per capita, constant 2010 US\\$, World Bank national accounts data, and OECD National Accounts data files](https://data.worldbank.org/indicator/NY.GDP.PCAP.KD)."
]
},
{
"cell_type": "code",
"execution_count": 15,
"metadata": {},
"outputs": [],
"source": [
"table_GDP_capita = import_from_World_Bank(\"NY.GDP.PCAP.KD\", \"GDP_per_capita\")\n",
"PISA = PISA.merge(table_GDP_capita, on=[\"iso3code\", \"Year\"], how=\"left\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 5. Espérance de vie à la naissance (source : Banque Mondiale)\n",
"\n",
"Source : [Life expectancy at birth, total (years), United Nations](https://data.worldbank.org/indicator/SP.DYN.LE00.IN)"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"# Ajoutons cette information à la base principale.\n",
"table_life_expect = import_from_World_Bank(\"SP.DYN.LE00.IN\", \"life_expect\")\n",
"PISA = PISA.merge(table_life_expect, on=[\"iso3code\", \"Year\"], how=\"left\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 6. Dépenses gouvernementales dans l'éducation, en % du PIB (source : Banque Mondiale)\n",
"\n",
"Source : [Government expenditure on education, total (% of GDP), UNESCO Institute for Statistics (uis.unesco.org)](https://data.worldbank.org/indicator/SE.XPD.TOTL.GD.ZS)"
]
},
{
"cell_type": "code",
"execution_count": 17,
"metadata": {},
"outputs": [],
"source": [
"# Ajoutons cette information à la base principale.\n",
"table_gov_exp = import_from_World_Bank(\"SE.XPD.TOTL.GD.ZS\", \"gov_exp\")\n",
"PISA = PISA.merge(table_gov_exp, on=[\"iso3code\", \"Year\"], how=\"left\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 7. Nombre d'élèves par enseignant à l'école primaire (source : Banque Mondiale)\n",
"\n",
"Source : [Pupil-teacher ratio, primary, UNESCO Institute for Statistics (uis.unesco.org)](https://data.worldbank.org/indicator/SE.PRM.ENRL.TC.ZS)"
]
},
{
"cell_type": "code",
"execution_count": 18,
"metadata": {},
"outputs": [],
"source": [
"table_pupil_teacher_ratio = import_from_World_Bank(\"SE.PRM.ENRL.TC.ZS\", \"pupil_teacher_ratio\")\n",
"PISA = PISA.merge(table_pupil_teacher_ratio, on=[\"iso3code\", \"Year\"], how=\"left\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### C. Finalisation de la base de données complète"
]
},
{
"cell_type": "code",
"execution_count": 19,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Country
\n",
"
Score
\n",
"
Subject
\n",
"
Year
\n",
"
is_OECD
\n",
"
iso3code
\n",
"
GDP_per_capita
\n",
"
life_expect
\n",
"
gov_exp
\n",
"
pupil_teacher_ratio
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
China
\n",
"
590
\n",
"
Science
\n",
"
2018
\n",
"
0
\n",
"
CHN
\n",
"
7752.559525
\n",
"
76.470000
\n",
"
1.88804
\n",
"
16.42675
\n",
"
\n",
"
\n",
"
1
\n",
"
Singapore
\n",
"
551
\n",
"
Science
\n",
"
2018
\n",
"
0
\n",
"
SGP
\n",
"
58247.872640
\n",
"
82.895122
\n",
"
2.89769
\n",
"
14.69428
\n",
"
\n",
"
\n",
"
2
\n",
"
Macau
\n",
"
544
\n",
"
Science
\n",
"
2018
\n",
"
0
\n",
"
MAC
\n",
"
58641.627664
\n",
"
83.989000
\n",
"
2.71071
\n",
"
13.49843
\n",
"
\n",
"
\n",
"
3
\n",
"
Estonia
\n",
"
530
\n",
"
Science
\n",
"
2018
\n",
"
1
\n",
"
EST
\n",
"
19954.130111
\n",
"
77.641463
\n",
"
5.17316
\n",
"
11.31153
\n",
"
\n",
"
\n",
"
4
\n",
"
Japan
\n",
"
529
\n",
"
Science
\n",
"
2018
\n",
"
1
\n",
"
JPN
\n",
"
48919.798942
\n",
"
84.099756
\n",
"
3.59059
\n",
"
15.66096
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Country Score Subject Year is_OECD iso3code GDP_per_capita \\\n",
"0 China 590 Science 2018 0 CHN 7752.559525 \n",
"1 Singapore 551 Science 2018 0 SGP 58247.872640 \n",
"2 Macau 544 Science 2018 0 MAC 58641.627664 \n",
"3 Estonia 530 Science 2018 1 EST 19954.130111 \n",
"4 Japan 529 Science 2018 1 JPN 48919.798942 \n",
"\n",
" life_expect gov_exp pupil_teacher_ratio \n",
"0 76.470000 1.88804 16.42675 \n",
"1 82.895122 2.89769 14.69428 \n",
"2 83.989000 2.71071 13.49843 \n",
"3 77.641463 5.17316 11.31153 \n",
"4 84.099756 3.59059 15.66096 "
]
},
"execution_count": 19,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Version définitive de la table, qui sera utilisée dans la suite de ce devoir.\n",
"PISA.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"De manière alternative, on propose une version \"pivotée\", qui pourra nous faciliter la tâche dans la suite."
]
},
{
"cell_type": "code",
"execution_count": 20,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Country
\n",
"
Year
\n",
"
is_OECD
\n",
"
iso3code
\n",
"
GDP_per_capita
\n",
"
life_expect
\n",
"
gov_exp
\n",
"
pupil_teacher_ratio
\n",
"
Maths
\n",
"
Reading
\n",
"
Science
\n",
"
Mean_score
\n",
"
\n",
" \n",
" \n",
"
\n",
"
394
\n",
"
Uruguay
\n",
"
2018
\n",
"
0
\n",
"
URY
\n",
"
14617.464002
\n",
"
77.632000
\n",
"
4.86891
\n",
"
11.01930
\n",
"
418.0
\n",
"
427.0
\n",
"
426.0
\n",
"
423.7
\n",
"
\n",
"
\n",
"
388
\n",
"
United States
\n",
"
2018
\n",
"
1
\n",
"
USA
\n",
"
54579.016837
\n",
"
78.539024
\n",
"
4.96174
\n",
"
14.19857
\n",
"
478.0
\n",
"
505.0
\n",
"
502.0
\n",
"
495.0
\n",
"
\n",
"
\n",
"
381
\n",
"
United Kingdom
\n",
"
2018
\n",
"
1
\n",
"
GBR
\n",
"
43324.592970
\n",
"
81.156098
\n",
"
5.48697
\n",
"
15.13275
\n",
"
502.0
\n",
"
504.0
\n",
"
505.0
\n",
"
503.7
\n",
"
\n",
"
\n",
"
374
\n",
"
United Arab Emirates
\n",
"
2018
\n",
"
0
\n",
"
ARE
\n",
"
40782.443624
\n",
"
77.647000
\n",
"
NaN
\n",
"
24.52278
\n",
"
435.0
\n",
"
432.0
\n",
"
434.0
\n",
"
433.7
\n",
"
\n",
"
\n",
"
371
\n",
"
Ukraine
\n",
"
2018
\n",
"
0
\n",
"
UKR
\n",
"
3110.194646
\n",
"
71.780976
\n",
"
5.41400
\n",
"
12.98011
\n",
"
453.0
\n",
"
466.0
\n",
"
469.0
\n",
"
462.7
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Country Year is_OECD iso3code GDP_per_capita \\\n",
"394 Uruguay 2018 0 URY 14617.464002 \n",
"388 United States 2018 1 USA 54579.016837 \n",
"381 United Kingdom 2018 1 GBR 43324.592970 \n",
"374 United Arab Emirates 2018 0 ARE 40782.443624 \n",
"371 Ukraine 2018 0 UKR 3110.194646 \n",
"\n",
" life_expect gov_exp pupil_teacher_ratio Maths Reading Science \\\n",
"394 77.632000 4.86891 11.01930 418.0 427.0 426.0 \n",
"388 78.539024 4.96174 14.19857 478.0 505.0 502.0 \n",
"381 81.156098 5.48697 15.13275 502.0 504.0 505.0 \n",
"374 77.647000 NaN 24.52278 435.0 432.0 434.0 \n",
"371 71.780976 5.41400 12.98011 453.0 466.0 469.0 \n",
"\n",
" Mean_score \n",
"394 423.7 \n",
"388 495.0 \n",
"381 503.7 \n",
"374 433.7 \n",
"371 462.7 "
]
},
"execution_count": 20,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"PISA_pivote = PISA\n",
"\n",
"# Avant de pivoter, on remplace les N/A par des fausses valeurs (seule solution trouvée pour éviter leur disparition)\n",
"PISA_pivote = PISA_pivote.fillna(\"foo\")\n",
"PISA_pivote = PISA_pivote.pivot_table(columns=\"Subject\", values=\"Score\", \n",
" index=[\"Country\", \"Year\", \"is_OECD\", \"iso3code\", \"GDP_per_capita\", \n",
" \"life_expect\", \"gov_exp\", \"pupil_teacher_ratio\"])\n",
"PISA_pivote = PISA_pivote.reset_index(drop=False).rename_axis(None, axis=1)\n",
"PISA_pivote = PISA_pivote.replace(\"foo\", np.nan)\n",
"\n",
"# On lui ajoute une colonne contenant le score moyen par pays et par année.\n",
"PISA_pivote[\"Mean_score\"] = PISA_pivote.apply(\n",
" lambda row: np.round(np.nanmean([row[\"Maths\"], row[\"Reading\"], row[\"Science\"]]),1), 1)\n",
"\n",
"PISA_pivote = PISA_pivote.sort_values(by=[\"Year\", \"Country\"], ascending=False)\n",
"PISA_pivote.head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On exporte à toutes fins utiles les deux versions de notre base de données au format csv."
]
},
{
"cell_type": "code",
"execution_count": 21,
"metadata": {},
"outputs": [],
"source": [
"PISA_pivote.to_csv(\"PISA_pivote.csv\")\n",
"PISA.to_csv(\"PISA.csv\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## II. Statistiques descriptives"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### A. Focus sur les résultats obtenus en 2018"
]
},
{
"cell_type": "code",
"execution_count": 22,
"metadata": {},
"outputs": [],
"source": [
"Specific_year = 2018\n",
"Ranking_specific_year = PISA_pivote[PISA_pivote[\"Year\"]==Specific_year]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 1. Qui sont les grands gagnants et les grands perdants en 2018 ? (selon le score moyen obtenu aux 3 compétences)"
]
},
{
"cell_type": "code",
"execution_count": 23,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Top 5 des meilleurs scores PISA en 2018 :\n"
]
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Country
\n",
"
Maths
\n",
"
Reading
\n",
"
Science
\n",
"
Mean_score
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
China
\n",
"
591.0
\n",
"
555.0
\n",
"
590.0
\n",
"
578.7
\n",
"
\n",
"
\n",
"
1
\n",
"
Singapore
\n",
"
569.0
\n",
"
549.0
\n",
"
551.0
\n",
"
556.3
\n",
"
\n",
"
\n",
"
2
\n",
"
Macau
\n",
"
558.0
\n",
"
525.0
\n",
"
544.0
\n",
"
542.3
\n",
"
\n",
"
\n",
"
3
\n",
"
Hong Kong
\n",
"
551.0
\n",
"
524.0
\n",
"
517.0
\n",
"
530.7
\n",
"
\n",
"
\n",
"
4
\n",
"
Estonia
\n",
"
523.0
\n",
"
523.0
\n",
"
530.0
\n",
"
525.3
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Country Maths Reading Science Mean_score\n",
"0 China 591.0 555.0 590.0 578.7\n",
"1 Singapore 569.0 549.0 551.0 556.3\n",
"2 Macau 558.0 525.0 544.0 542.3\n",
"3 Hong Kong 551.0 524.0 517.0 530.7\n",
"4 Estonia 523.0 523.0 530.0 525.3"
]
},
"execution_count": 23,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"print(\"Top 5 des meilleurs scores PISA en 2018 :\")\n",
"Ranking_specific_year[['Country', 'Maths', 'Reading', 'Science', 'Mean_score']].sort_values(by=\"Mean_score\",ascending=False).reset_index(drop=True).head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Remarques sur ce top 5 :\n",
"* Pour la **Chine**, le test PISA ne couvre en fait que les villes de Beijing, Shanghai, Jiangsu et Zhejiang. Sa performance exceptionnelle est donc à prendre avec des pincettes. Il n'en demeure pas moins que, sur ces régions, son évolution est épatante puisque le pays a gagné 64 points en moyenne depuis 2015.\n",
"* On remarque que les pays qui atteignent le haut du classement sont pour la plupart portés par les disciplines scientifiques : le cas de la **Chine** est emblématique puisque son score en lecture est inférieur d'environ 35 points à son score en mathématiques et en sciences. Dans ce top 5, aucun pays n'obtient un meilleur score en lecture qu'en mathématiques ou en sciences.\n",
"* Seul le **Japon** voit son score moyen diminuer par rapport à l'édition 2015 : en fait, le pays a perdu des points dans toutes les composantes du test PISA. Son score est en fait très décevant si l'on s'intéresse à la tendance.\n",
"\n",
"Intéressons-nous désormais aux 5 pays obtenant les moins bons scores en 2018."
]
},
{
"cell_type": "code",
"execution_count": 24,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Top 5 des pires scores PISA en 2018 :\n"
]
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Country
\n",
"
Maths
\n",
"
Reading
\n",
"
Science
\n",
"
Mean_score
\n",
"
\n",
" \n",
" \n",
"
\n",
"
0
\n",
"
Dominican Republic
\n",
"
325.0
\n",
"
342.0
\n",
"
336.0
\n",
"
334.3
\n",
"
\n",
"
\n",
"
1
\n",
"
Philippines
\n",
"
353.0
\n",
"
340.0
\n",
"
357.0
\n",
"
350.0
\n",
"
\n",
"
\n",
"
2
\n",
"
Kosovo
\n",
"
366.0
\n",
"
353.0
\n",
"
365.0
\n",
"
361.3
\n",
"
\n",
"
\n",
"
3
\n",
"
Panama
\n",
"
353.0
\n",
"
377.0
\n",
"
365.0
\n",
"
365.0
\n",
"
\n",
"
\n",
"
4
\n",
"
Morocco
\n",
"
368.0
\n",
"
359.0
\n",
"
377.0
\n",
"
368.0
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Country Maths Reading Science Mean_score\n",
"0 Dominican Republic 325.0 342.0 336.0 334.3\n",
"1 Philippines 353.0 340.0 357.0 350.0\n",
"2 Kosovo 366.0 353.0 365.0 361.3\n",
"3 Panama 353.0 377.0 365.0 365.0\n",
"4 Morocco 368.0 359.0 377.0 368.0"
]
},
"execution_count": 24,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"print(\"Top 5 des pires scores PISA en 2018 :\")\n",
"Ranking_specific_year[['Country', 'Maths', 'Reading', 'Science', 'Mean_score']].sort_values(by=\"Mean_score\",ascending=True).reset_index(drop=True).head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On observe que les pays obtenant les pires scores moyens ont quasiment tous (sauf le Liban, qui stagne) perdu des places par rapport à l'édition 2015."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 2. Comment sont réparties les performances en 2018 ?\n",
"On va se demander à quoi ressemble un \"bon\" ou un \"mauvais\" score PISA en 2018 en étudiant la répartition des scores parmi les pays participants."
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
count
\n",
"
mean
\n",
"
std
\n",
"
min
\n",
"
25%
\n",
"
50%
\n",
"
75%
\n",
"
max
\n",
"
\n",
" \n",
" \n",
"
\n",
"
Maths
\n",
"
78.0
\n",
"
458.6
\n",
"
56.4
\n",
"
325.0
\n",
"
417.2
\n",
"
468.0
\n",
"
500.0
\n",
"
591.0
\n",
"
\n",
"
\n",
"
Reading
\n",
"
77.0
\n",
"
453.1
\n",
"
53.1
\n",
"
340.0
\n",
"
412.0
\n",
"
466.0
\n",
"
498.0
\n",
"
555.0
\n",
"
\n",
"
\n",
"
Science
\n",
"
78.0
\n",
"
457.9
\n",
"
51.9
\n",
"
336.0
\n",
"
417.5
\n",
"
468.0
\n",
"
498.5
\n",
"
590.0
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" count mean std min 25% 50% 75% max\n",
"Maths 78.0 458.6 56.4 325.0 417.2 468.0 500.0 591.0\n",
"Reading 77.0 453.1 53.1 340.0 412.0 466.0 498.0 555.0\n",
"Science 78.0 457.9 51.9 336.0 417.5 468.0 498.5 590.0"
]
},
"execution_count": 25,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"np.round(Ranking_specific_year[[\"Maths\", \"Reading\", \"Science\"]].describe(),1).transpose()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On remarque que, pour l'année 2018 :\n",
"* **78 pays** ont pris part aux tests PISA (sauf en lecture, où le score de l'Espagne n'est pas référencé).\n",
"* Le **score moyen** est très proche dans les trois disciplines : il est de 453 en lecture, 458 en science, 459 en mathématiques.\n",
"* Le **score médian** est également très proche dans les trois disciplines : il est de 466 en lecture, 468 en science, 468 en mathématiques.\n",
"* Les scores sont tous compris **entre 325 et 591**. À noter que les mathématiques sont la matière la plus discriminante : \n",
" * C'est dans cette discipline que l'écart-type est le plus élevé (56 contre 53 et 52 pour la lecture et les sciences).\n",
" * C'est aussi dans cette compétence que les scores descendent le plus bas et montent le plus haut.\n",
"\n",
"Intéressons-nous désormais à la distribution des scores."
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.set()\n",
"fig, axs = plt.subplots(1, 2, figsize=(17,5))\n",
"\n",
"sns.distplot(Ranking_specific_year[\"Maths\"].dropna(), hist=False, label=\"Maths\", ax = axs[0])\n",
"sns.distplot(Ranking_specific_year[\"Reading\"].dropna(), hist=False, label=\"Reading\", ax = axs[0])\n",
"sns.distplot(Ranking_specific_year[\"Science\"].dropna(), hist=False, label=\"Science\", ax = axs[0])\n",
"sns.distplot(Ranking_specific_year[\"Mean_score\"].dropna(), hist=False, kde=False, rug=True, ax = axs[0])\n",
"axs[0].set_title(\"Fonction de densité des scores\")\n",
"axs[0].set_xlabel(\"Score obtenu à l'épreuve\")\n",
"\n",
"sns.boxplot(x=\"Subject\", y=\"Score\", data=PISA[PISA[\"Year\"]==Specific_year], palette=\"Set3\", ax = axs[1])\n",
"axs[1].set_title(\"Répartition du score dans chaque discipline lors de l'édition 2018 du test PISA\")\n",
"axs[1].set_ylabel(\"Score obtenu à l'épreuve\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On constate que :\n",
"* La répartition des scores est assez proche pour les trois compétences testées.\n",
"* 50% des pays participants ont un score compris entre environ 420 et 500 pour les trois disciplines.\n",
"* Les trois densités s'apparentent à une somme de deux courbes en cloche, l'une qui atteindrait son pic aux alentours de 430, l'autre qui atteindrait son pic aux alentours de 490. Il semble donc, on y reviendra plus loin, que les pays participants peuvent être séparés en deux sous-groupes assez homogènes.\n",
"* Les valeurs sont un peu plus faibles et un peu plus concentrées pour l'épreuve de lecture que pour les épreuves de maths et de sciences."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 3. Quels sont les pays dont le score moyen a le plus évolué entre 2015 et 2018 ?"
]
},
{
"cell_type": "code",
"execution_count": 27,
"metadata": {},
"outputs": [],
"source": [
"Ranking_previous_test = PISA_pivote[PISA_pivote[\"Year\"]==Specific_year - 3][[\"Country\", \"Mean_score\"]]\n",
"Ranking_previous_test.columns = [\"Country\", \"Mean_score_previous_test\"]\n",
"Ranking_specific_year = Ranking_specific_year.merge(Ranking_previous_test, on=\"Country\")\n",
"new_col = Ranking_specific_year.apply(lambda row : (row[\"Mean_score\"]-row[\"Mean_score_previous_test\"]), axis=1)\n",
"Ranking_specific_year[\"Evolution\"] = new_col"
]
},
{
"cell_type": "code",
"execution_count": 28,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
" \n",
" "
]
},
"metadata": {},
"output_type": "display_data"
},
{
"data": {
"application/vnd.plotly.v1+json": {
"config": {
"plotlyServerURL": "https://plot.ly"
},
"data": [
{
"coloraxis": "coloraxis",
"geo": "geo",
"hoverlabel": {
"namelength": 0
},
"hovertemplate": "%{hovertext}
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = px.choropleth(Ranking_specific_year, locations=\"iso3code\",\n",
" color=\"Evolution\", hover_name=\"Country\",\n",
" color_continuous_scale=\"Spectral\", animation_frame=\"Year\",\n",
" range_color = [-30,30],\n",
" title='Évolution du score obtenu en 2018 par rapport à celui obtenu en 2015 (moyenne sur les 3 disciplines)')\n",
"fig.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"L'édition 2018 révèle une tendance préoccupante en Europe particulièrement où la majorité des pays voient leur score diminuer par rapport à l'édition 2015. Remarquons la présence de quelques valeurs extrêmes :\n",
"* Du côté des très bonnes surprises, on retrouve comme attendu la **Chine** (+64), mais aussi la **Turquie** (+38), la **Jordanie** (+17).\n",
"* Du côté des très mauvaises surprises, on trouve le **Kazakhstan** (-45), l'**Argentine** (-27), la **Géorgie** (-18)."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### B. Évolution des résultats par année et par pays"
]
},
{
"cell_type": "code",
"execution_count": 29,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Utiliser la barre de temps en bas de la carte pour faire défiler les années.\n"
]
},
{
"data": {
"application/vnd.plotly.v1+json": {
"config": {
"plotlyServerURL": "https://plot.ly"
},
"data": [
{
"coloraxis": "coloraxis",
"geo": "geo",
"hoverlabel": {
"namelength": 0
},
"hovertemplate": "%{hovertext}
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = px.choropleth(PISA_pivote, locations=\"iso3code\",\n",
" color=\"Mean_score\", hover_name=\"Country\",\n",
" color_continuous_scale=\"Spectral\", animation_frame=\"Year\",\n",
" range_color = [350,550],\n",
" title='Score obtenu au test PISA par année et par pays (moyenne sur les 3 disciplines)')\n",
"\n",
"print(\"Utiliser la barre de temps en bas de la carte pour faire défiler les années.\")\n",
"fig.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Rappel : pour la Chine, le test PISA ne couvre en fait que les villes de Beijing, Shanghai, Jiangsu et Zhejiang. On affiche la valeur sur tout le territoire par abus, mais celle-ci est donc à considérer avec précaution.\n",
"\n",
"Quelques cas intéressants révélés par cette carte, en guise d'illustration :\n",
"* On a vu ci-dessus que le **Kazakhstan** était le pays dont le score moyen a le plus reculé en 2018 par rapport à 2015. Cette contre-performance est particulièrement surprenante si l'on s'intéresse au chemin parcouru, car elle s'inscrit à contre-courant d'une dynamique très favorable. Depuis 2009, son score ne cessait de croître (passant de 398 en 2009 à 448 en 2015), mais tous les progrès ont vraisemblablement disparu en trois ans (402 en 2018).\n",
"* On a vu ci-dessus que la **Turquie** avait gagné 38 points entre 2015 et 2018, obtenant un score moyen de 463. En fait, il ne s'agit pas d'une dynamique de long-terme mais plutôt de la correction d'un choc brutal survenu en 2015. Le pays avait déjà un score moyen de 462 en 2012. Il avait chuté soudainement à 424 en 2015. Le score de 2018 n'est donc qu'un retour à la situation connue en 2012.\n",
"* L'**Allemagne** a souvent été citée comme un cas emblématique, car le pays aurait vécu un véritable [\"choc PISA\"](https://www.lemonde.fr/education/article/2019/12/03/en-allemagne-l-enquete-pisa-a-provoque-un-sursaut-du-systeme-educatif_6021463_1473685.html) en découvrant, dès 2000, que son score moyen était bien inférieur à la moyenne des pays de l'OCDE. Elle aurait alors entrepris d'importantes mesures afin d'améliorer le niveau de ses élèves. En effet : son score moyen est passé de 484 en 2000 à 515 en 2012. Mais depuis 2015, il chute : il repasse à 500 pour l'édition 2018 du test PISA. Le pays a donc perdu la moitié des points qu'il avait réussi à gagner en 12 ans...\n",
"* L'Europe de l'Est a connu un phénomène de rattrapage au début des années 2000, qui s'est là aussi interrompu en 2012. Par exemple, la **Pologne** avait un score de 479 en 2000. Il croît constamment jusqu'en 2012 où il atteint 521, mais il décroche ensuite et retombe à 513.\n",
"* Les pays scandinaves sont souvent évoqués pour la réussite de leur système scolaire. Pourtant, leurs performances au test PISA sont mitigées. Dans l'absolu, leurs scores sont très élevés, mais leur dynamique n'est pas des plus favorables. Le score moyen de la **Finlande** était très élevé (553) en 2006, mais il chute incessamment depuis (516 en 2018)."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### C. Comment va la France ?"
]
},
{
"cell_type": "code",
"execution_count": 30,
"metadata": {},
"outputs": [],
"source": [
"PISA_France = PISA[PISA[\"Country\"]==\"France\"]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 1. Évolution de sa performance au fil du temps"
]
},
{
"cell_type": "code",
"execution_count": 31,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.set()\n",
"fig, axs = plt.subplots(1, 1, figsize=(17,5))\n",
"sns.lineplot(x=\"Year\", y=\"Score\", hue=\"Subject\", data=PISA_France, ax=axs, linewidth=2)\n",
"plt.xticks(range(2000,2021,3))\n",
"plt.title(\"Évolution des performances de la France en lecture, mathématiques et sciences depuis 2000.\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On constate, comme l'ont souligné de nombreux journaux, que la France voit ses performances diminuer régulièrement depuis 2012. Plus particulièrement : dans la dernière enquête PISA réalisée en 2018, la France a perdu 6 points en lecture et 2 en sciences, en comparaison avec la précédente étude réalisée en 2015. Seule consolation : elle en regagne 2 en mathématiques."
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Country
\n",
"
Year
\n",
"
Mean_score
\n",
"
\n",
" \n",
" \n",
"
\n",
"
115
\n",
"
France
\n",
"
2018
\n",
"
493.7
\n",
"
\n",
"
\n",
"
114
\n",
"
France
\n",
"
2015
\n",
"
495.7
\n",
"
\n",
"
\n",
"
113
\n",
"
France
\n",
"
2012
\n",
"
499.7
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Country Year Mean_score\n",
"115 France 2018 493.7\n",
"114 France 2015 495.7\n",
"113 France 2012 499.7"
]
},
"execution_count": 32,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"PISA_France_pivote = PISA_pivote[PISA_pivote[\"Country\"]==\"France\"]\n",
"PISA_France_pivote[PISA_France_pivote[\"Year\"].isin(range(2012,2019))][[\"Country\", \"Year\", \"Mean_score\"]]"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"En moyenne, toutes disciplines confondes, elle a ainsi perdu 4 poids entre 2012 et 2015, puis encore 2 points entre 2015 et 2018."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 2. Comparaison avec la performance moyenne des pays de l'OCDE\n",
"Cela n'aurait pas de sens de comparer la France avec la performance moyenne de tous les pays participants aux tests PISA car **de nouveaux pays** (souvent, des pays aux performances inférieures à la moyenne) **s'ajoutent à chaque édition**. On choisit donc de comparer la France aux pays membres de l'OCDE, qui constituent un ensemble de pays relativement homogène, qui participent historiquement à ce test, et qui sont souvent pris pour groupe de référence."
]
},
{
"cell_type": "code",
"execution_count": 33,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"# Calcul des scores moyens pour les pays de l'OCDE\n",
"PISA_OECD_aggregated = PISA[PISA[\"is_OECD\"]==1].groupby(['Year', 'Subject', 'is_OECD']).agg(\"mean\")\n",
"PISA_OECD_aggregated = PISA_OECD_aggregated.reset_index(drop=False)\n",
"PISA_OECD_aggregated[\"Country\"] = \"OECD mean\"\n",
"PISA_France_vs_OECD = pd.concat([PISA_France, PISA_OECD_aggregated], sort = True)\n",
"\n",
"# Affichage du graphique comparant la France avec les pays de l'OCDE\n",
"sns.set()\n",
"fig, axs = plt.subplots(1, 1, figsize=(17,5))\n",
"sns.lineplot(x=\"Year\", y=\"Score\", hue=\"Subject\", style=\"Country\", data=PISA_France_vs_OECD, ax=axs, linewidth=2)\n",
"plt.xticks(range(2000,2021,3))\n",
"plt.title(\"Évolution de la performance moyenne des pays de l'OCDE (pointillés) et de la France (trait plein) en lecture, mathématiques et sciences depuis 2000.\")\n",
"plt.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Ce graphique conduit à relativiser, ou du moins à donner un contexte, à la contre-performance française. Certes, la performance de la France ces dernières années n'est pas fameuse. Mais si on la compare, discipline par discipline, **la France est toujours au-dessus de la moyenne des pays de l'OCDE**.\n",
"* En lecture et en maths, la France suit une tendance très proche de celle de l'OCDE :\n",
" * Quand la performance des pays de l'OCDE augmente en moyenne dans ces deux disciplines (2006-2012), il en va de même pour la France.\n",
" * Quand la performance des pays de l'OCDE diminue en moyenne dans ces deux disciplines (2012-2018), il en va de même pour la France.\n",
"* En sciences, la France parvient à faire mieux que la tendance des pays de l'OCDE et à s'en \"libérer\" : \n",
" * En 2006-2012, elle est légèrement au-dessous de la moyenne de l'OCDE en sciences. \n",
" * Depuis 2015, elle fait mieux que la moyenne : si son niveau moyen dans cette discipline diminue, il diminue en fait bien moins vite que celui des autres pays de l'OCDE.\n",
"\n",
"Ce graphique pousse donc à tirer **des conclusions plus positives que précédemment** : en fait, tout dépend comment on utilise le score PISA.\n",
"* Si on le considère en tant que tel, alors oui, indéniablement, **le niveau des élèves français baisse constamment depuis 2012, surtout en lecture**. \n",
"* Si on le considère à la lumière de la performance moyenne des autres pays de l'OCDE, alors on peut se \"réjouir\" du fait que **la France ne soit pas seule dans cette situation** et qu'elle **demeure ainsi au-dessus de la moyenne**.\n",
"\n",
"Le commentateur, suivant son objectif, ne manquera pas d'insister sur l'un ou l'autre des constats pour \"faire parler les chiffres\". Sans surprise, le communiqué du Ministère de l'Éducation Nationale, cité en introduction, se concentre plutôt sur le second point."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### D. Évolution de la répartition de la performance au sein de l'OCDE\n",
"\n",
"Le package Plotly nous permet de proposer des boxplots dynamiques, qui facilitent la comparaison de la répartition au fil des ans."
]
},
{
"cell_type": "code",
"execution_count": 34,
"metadata": {},
"outputs": [],
"source": [
"PISA_OECD = PISA[PISA[\"is_OECD\"]==1].copy()\n",
"PISA_OECD = PISA_OECD[PISA_OECD[\"Year\"]>=2006]"
]
},
{
"cell_type": "code",
"execution_count": 35,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Utiliser la barre de temps en bas du graphique pour faire défiler les années.\n"
]
},
{
"data": {
"application/vnd.plotly.v1+json": {
"config": {
"plotlyServerURL": "https://plot.ly"
},
"data": [
{
"alignmentgroup": "True",
"boxpoints": "all",
"hoverlabel": {
"namelength": 0
},
"hovertemplate": "%{hovertext}
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = px.box(PISA_OECD, x=\"Subject\", y=\"Score\", hover_name=\"Country\", animation_frame=\"Year\", points=\"all\",\n",
" title=\"Évolution de la répartition des scores au sein de l'OCDE, par année et par compétence.\")\n",
"fig.update_yaxes(range=[400, 600])\n",
"print(\"Utiliser la barre de temps en bas du graphique pour faire défiler les années.\")\n",
"fig.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Deux constats principaux surgissent de ces représentations lorsqu'on fait varier l'axe temps. Entre 2012 et 2018 : \n",
"* **La performance baisse dans tous les domaines** : les principaux quantiles diminuent tous.\n",
"* **La performance s'homogénéise** :\n",
" * Les \"très bons\" deviennent progressivement moins bons. Par exemple, en 2012, les trois meilleurs pays de l'OCDE en sciences étaient le Japon (547), l'Estonie (545) et la Finlande (541). Ces trois pays ont vu leur score chuter depuis lors, passant respectivement à 529, 530 et 522. \n",
" * En revanche, les \"moins bons\" de l'OCDE tendent à stagner voire à reculer davantage, comme c'est le cas du Mexique et du Chili qui ne parviennent pas à surmonter leurs difficultés en maths."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## III. Approfondissements\n",
"Cette partie se concentre sur les tests réalisés après 2006, car toutes les épreuves n'existaient pas encore les années précédentes."
]
},
{
"cell_type": "code",
"execution_count": 36,
"metadata": {},
"outputs": [],
"source": [
"PISA_pivote_post2006 = PISA_pivote[PISA_pivote[\"Year\"]>=2006].copy()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### A. Quelle corrélation entre le score dans une compétence et une autre, pour un pays et une année donnés ?"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On est donc en présence, pour chaque pays, et pour chaque édition du test PISA, de trois variables : le score en maths, en lecture, et en sciences. On a pu observer, sur divers exemples, que la majorité des pays obtenaient des scores très similaires dans les trois domaines (un pays \"bon\" en maths est généralement \"bon\" en sciences et en lecture). De plus, on a étudié, à plusieurs reprises, le \"score moyen\" calculé par pays et par année. Cet indicateur est pratique parce qu'il facilite la comparaison, mais il peut néanmoins cacher des disparités. On voudrait donc savoir s'il existe des pays qui performent particulièrement mieux dans un domaine que dans un autre, ce que ne permettrait pas de voir un score moyen."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 1. Première approche : corrélation entre les 3 variables PISA"
]
},
{
"cell_type": "code",
"execution_count": 37,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Matrice de corrélation entre les scores obtenus dans les trois compétences.\n"
]
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Maths
\n",
"
Reading
\n",
"
Science
\n",
"
\n",
" \n",
" \n",
"
\n",
"
Maths
\n",
"
1.000
\n",
"
0.949
\n",
"
0.974
\n",
"
\n",
"
\n",
"
Reading
\n",
"
0.949
\n",
"
1.000
\n",
"
0.969
\n",
"
\n",
"
\n",
"
Science
\n",
"
0.974
\n",
"
0.969
\n",
"
1.000
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" Maths Reading Science\n",
"Maths 1.000 0.949 0.974\n",
"Reading 0.949 1.000 0.969\n",
"Science 0.974 0.969 1.000"
]
},
"execution_count": 37,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"print(\"Matrice de corrélation entre les scores obtenus dans les trois compétences.\")\n",
"np.round(PISA_pivote_post2006[[\"Maths\", \"Reading\", \"Science\"]].corr(),3)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On constate sans surprise que les trois variables sont très fortement corrélées :\n",
"* Maths et Reading sont corrélées à 94,9%\n",
"* Maths et Science sont corrélées à 97,3%\n",
"* Science et Reading sont corrélées à 96,9%\n",
"\n",
"Obtenir un bon ou un mauvais score dans une compétence est donc fortement corrélé au fait d'obtenir un bon ou un mauvais score dans une autre compétence. Ce résultat est confirmé par les graphiques croisés ci-dessous, qui croisent le score dans une matière avec le score dans une autre matière."
]
},
{
"cell_type": "code",
"execution_count": 38,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Score dans une matière en fonction du score dans une autre matière.\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.pairplot(PISA_pivote_post2006[[\"Maths\", \"Reading\", \"Science\"]].dropna(), diag_kind=\"kde\")\n",
"print(\"Score dans une matière en fonction du score dans une autre matière.\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 2. Deuxième approche : calcul des écarts [max-min]\n",
"On constate que certains pays s'éloignent quand même légèrement de la bissectrice sur les graphiques ci-dessus. Cela est particulièrement visible sur le graphique qui croise la performance en maths et en lecture, les deux composantes les moins corrélées. On voudrait pouvoir identifier ces pays.\n",
"Pour cela, on va étudier l'écart [max(maths, reading, science) - min(maths, reading, science)], pour chaque pays et pour chaque année."
]
},
{
"cell_type": "code",
"execution_count": 39,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Top 5 des pays ayant le plus gros écart [max-min] en 2018\n"
]
},
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = px.scatter_3d(PISA_pivote_post2006, x=\"Maths\", y=\"Reading\", z=\"Science\", color=\"Ecart_max\", \n",
" hover_name=\"Country\", hover_data=[\"Maths\", \"Reading\", \"Science\", \"Mean_score\"], \n",
" color_continuous_scale=\"Picnic\", animation_frame=\"Year\", range_color = [0,40],\n",
" title=\"Visualisation 3D du score en maths, lecture, sciences, par pays et par année.\")\n",
"\n",
"fig.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 4. Que retenir ?\n",
"Les considérations ci-dessus incitent donc à la vigilance quand on parle du \"classement\" PISA : il est pratique de comparer des scores moyens, comme nous l'avons fait, et comme il est courant de le faire par ailleurs. Les moyennes demeurent utiles et pertinentes, parce que les écarts restent généralement modérés : il n'y a pas de pays qui serait premier en mathématiques et dernier en lecture, par exemple... Toutefois, se limiter à des moyennes est une approche incomplète. Par exemple, les Pays-Bas obtiennent un score moyen supérieur de 10 points à celui de la France. On serait alors tentés d'en conclure que ce pays est \"meilleur\", \"mieux classé\", ou même que son système scolaire fonctionne mieux. Pourtant, ses résultats en lecture sont inférieurs de 8 points à ceux de la France (485 vs 493). On retiendra donc qu'il est indispensable d'adopter quelques précautions de langage quand on parle de \"classement PISA\" et que les moyennes ne sont qu'un agrégat qui cache des disparités."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### B. Formation de clusters par méthode de K-means\n",
"\n",
"Dans la partie [II - A - 2. Comment sont réparties les performances en 2018 ?](#2.-Comment-sont-réparties-les-performances-en-2018-?), on a représenté la densité des scores obtenus en maths, en lecture et en sciences en 2018. On a remarqué, en observant ces courbes, qu'elles s'apparentaient \"à vue d'oeil\", à la somme de deux courbes en cloche, ce qui laisse à penser qu'on pourrait scinder les pays participants en deux groupes relativement homogènes en termes de performance. Au moyen d'un algorithme k-means, on va chercher à classifier les pays participants au test PISA en deux catégories, pour chaque édition du test."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 1. Formation des clusters pour chaque année\n",
"Les seules variables d'entrée utilisées pour former les clusters sont le score en mathématiques, en lecture et en science. Les différents indicateurs économiques qui seront étudiés plus tard n'entrent pas du tout en compte dans la formation des clusters."
]
},
{
"cell_type": "code",
"execution_count": 42,
"metadata": {},
"outputs": [],
"source": [
"sub_table = PISA_pivote[pd.notnull(PISA_pivote[\"Maths\"]) & pd.notnull(PISA_pivote[\"Reading\"]) \n",
" & pd.notnull(PISA_pivote[\"Science\"])].copy()\n",
"\n",
"model=KMeans(n_clusters=2)\n",
"PISA_pivote_clusters = pd.DataFrame()\n",
"\n",
"for Year in sub_table[\"Year\"].unique():\n",
" \n",
" # On calcule les clusters séparément pour chaque année du test PISA.\n",
" sub_table_year = sub_table[sub_table[\"Year\"] == Year].copy()\n",
" model.fit(sub_table_year[[\"Maths\", \"Science\", \"Reading\"]])\n",
" sub_table_year[\"Cluster\"] = model.labels_\n",
" \n",
" # Précaution pour s'assurer que les id de clusters soient rangés par niveau de performance moyen croissant.\n",
" Mean_by_cluster = sub_table_year.groupby(\"Cluster\", as_index=False).agg(\"mean\").sort_values(by=\"Mean_score\")\n",
" i=1\n",
" for Cluster in Mean_by_cluster['Cluster']:\n",
" sub_table_year['Cluster'] = sub_table_year['Cluster'].replace(Cluster, 'Cluster '+str(i))\n",
" i+=1\n",
" \n",
" PISA_pivote_clusters = pd.concat([PISA_pivote_clusters, sub_table_year])"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 2. Visualisation des individus composant chaque cluster, par année"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [
{
"data": {
"application/vnd.plotly.v1+json": {
"config": {
"plotlyServerURL": "https://plot.ly"
},
"data": [
{
"customdata": [
[
418,
427,
426
],
[
435,
432,
434
],
[
419,
393,
426
],
[
448,
439,
440
],
[
373,
399,
386
],
[
430,
428,
426
],
[
414,
407,
419
],
[
353,
340,
357
],
[
400,
401,
404
],
[
353,
377,
365
],
[
394,
393,
413
],
[
368,
359,
377
],
[
430,
421,
415
],
[
421,
424,
428
],
[
440,
415,
438
],
[
393,
353,
384
],
[
366,
353,
365
],
[
423,
387,
397
],
[
400,
419,
429
],
[
379,
371,
396
],
[
398,
380,
383
],
[
325,
342,
336
],
[
451,
424,
439
],
[
402,
426,
416
],
[
391,
412,
413
],
[
436,
420,
424
],
[
430,
408,
431
],
[
384,
413,
404
],
[
406,
403,
398
],
[
420,
389,
398
],
[
379,
402,
404
],
[
437,
405,
417
]
],
"hoverlabel": {
"namelength": 0
},
"hovertemplate": "%{hovertext}
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = px.scatter_3d(PISA_pivote_clusters, x=\"Maths\", y=\"Reading\", z=\"Science\", color=\"Cluster\", \n",
" hover_name=\"Country\", hover_data=[\"Maths\", \"Reading\", \"Science\"], animation_frame=\"Year\",\n",
" title=\"Performance en maths, lecture et sciences des pays composant les deux clusters, par année.\",\n",
" symbol=\"is_OECD\")\n",
"\n",
"fig.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 3. Comparaison des deux clusters\n",
"Cette sous-partie se concentre sur l'année 2018 en guise d'illustration."
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Le cluster 1 est composé de 34 pays, dont le score moyen est compris entre 334.3 et 442.3\n",
"Le cluster 2 est composé de 43 pays, dont le score moyen est compris entre 453.3 et 578.7\n"
]
}
],
"source": [
"PISA_pivote_2018_clusters = PISA_pivote_clusters[PISA_pivote_clusters[\"Year\"]==2018].copy()\n",
"\n",
"Cluster_1 = PISA_pivote_2018_clusters[PISA_pivote_2018_clusters[\"Cluster\"]=='Cluster 1']\n",
"print(\"Le cluster 1 est composé de\", len(Cluster_1[\"iso3code\"]), \"pays, dont le score moyen est compris entre\", min(Cluster_1[\"Mean_score\"]), \"et\", max(Cluster_1[\"Mean_score\"]))\n",
"\n",
"Cluster_2 = PISA_pivote_2018_clusters[PISA_pivote_2018_clusters[\"Cluster\"]=='Cluster 2']\n",
"print(\"Le cluster 2 est composé de\", len(Cluster_2[\"iso3code\"]), \"pays, dont le score moyen est compris entre\", min(Cluster_2[\"Mean_score\"]), \"et\", max(Cluster_2[\"Mean_score\"]))\n"
]
},
{
"cell_type": "code",
"execution_count": 45,
"metadata": {},
"outputs": [
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"PISA_2018_clusters = PISA_pivote_2018_clusters.melt(id_vars=['Country', 'Year', 'is_OECD', 'iso3code', 'GDP_per_capita', 'life_expect', 'gov_exp', 'pupil_teacher_ratio', 'Cluster', 'Mean_score'], var_name='Subject', value_name='Score')\n",
"g = sns.FacetGrid(PISA_2018_clusters, row=\"Subject\", col=\"Cluster\", height=1.7, aspect=4)\n",
"g.map(sns.distplot, \"Score\", hist=False, rug=True);"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"On a bien réussi à décomposer les participants en deux catégories homogènes, dont les scores ont pour densité une courbe en cloche comme anticipé. Comparons les caractéristiques de ces deux clusters."
]
},
{
"cell_type": "code",
"execution_count": 46,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Comparaison des caractéristiques moyennes entre les deux clusters.\n"
]
},
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
"
is_OECD
\n",
"
GDP_per_capita
\n",
"
life_expect
\n",
"
gov_exp
\n",
"
pupil_teacher_ratio
\n",
"
Maths
\n",
"
Reading
\n",
"
Science
\n",
"
Mean_score
\n",
"
\n",
"
\n",
"
Cluster
\n",
"
Year
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
\n",
" \n",
" \n",
"
\n",
"
Cluster 1
\n",
"
2018
\n",
"
0.1
\n",
"
12324.5
\n",
"
76.0
\n",
"
4.2
\n",
"
17.3
\n",
"
404.2
\n",
"
401.0
\n",
"
407.4
\n",
"
404.2
\n",
"
\n",
"
\n",
"
Cluster 2
\n",
"
2018
\n",
"
0.7
\n",
"
39642.5
\n",
"
80.2
\n",
"
4.9
\n",
"
13.6
\n",
"
501.2
\n",
"
494.3
\n",
"
497.3
\n",
"
497.6
\n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" is_OECD GDP_per_capita life_expect gov_exp \\\n",
"Cluster Year \n",
"Cluster 1 2018 0.1 12324.5 76.0 4.2 \n",
"Cluster 2 2018 0.7 39642.5 80.2 4.9 \n",
"\n",
" pupil_teacher_ratio Maths Reading Science Mean_score \n",
"Cluster Year \n",
"Cluster 1 2018 17.3 404.2 401.0 407.4 404.2 \n",
"Cluster 2 2018 13.6 501.2 494.3 497.3 497.6 "
]
},
"execution_count": 46,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"print(\"Comparaison des caractéristiques moyennes entre les deux clusters.\")\n",
"np.round(PISA_pivote_2018_clusters.groupby(['Cluster', 'Year']).agg('mean'), 1)"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"* Le cluster 1 est composé des pays ayant un **score moyen supérieur à 450** : ils sont...\n",
" * Majoritairement membres de l'OCDE (72% d'entre eux)\n",
" * Majoritairement riches (PIB par tête de 39k \\\\$, de l'ordre de celui de la France qui est de 44k \\$)\n",
" * Investissent davantage dans l'éducation que la moyenne des participants (4,9% du PIB)\n",
" * Ont une espérance de vie à la naissance plus élevée que la moyenne (80 ans)\n",
" * Ont un ratio élèves par professeur à l'école primaire inférieur à la moyenne (13,6).\n",
"* Le cluster 2 est composé des pays ayant un **score moyen inférieur à 450** : ils sont...\n",
" * Majoritairement non membres de l'OCDE (6% d'entre eux)\n",
" * Majoritairement pauvres (PIB par tête de 12k \\$, 3 fois plus faible que dans le cluster 1)\n",
" * Investissent moins dans l'éducation que la moyenne des participants (4,2% du PIB)\n",
" * Ont une espérance de vie à la naissance plus faible que la moyenne (76 ans)\n",
" * Ont un ratio élèves par professeur à l'école primaire supérieur à la moyenne (17,3)."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 4. Que retenir ?\n",
"On observe donc l'existence de **deux groupes, tous deux plutôt homogènes** dans leur composition en termes de performance scolaire. Cette dichotomie semble assez intuitive puisqu'elle consiste à séparer :\n",
"* Des pays développés, riches, où l'on vit plus longtemps, où les gouvernements investissent davantage dans l'éducation, et où les performances scolaires sont plus élevées.\n",
"* Des pays moins développés, plus pauvres, où l'on vit moins longtemps, où les gouvernements investissent moins dans l'éducation, et où les performances scolaires sont moins élevées.\n",
"\n",
"Au passage, rappelons que tous les pays ne participent pas au test PISA : cette démarche est facultative, ce qui entraîne forcément un **biais d'auto-sélection**, et donc notre clusterisation ne permet pas de décrire l'état du monde mais seulement de caractériser un pool de pays bien particulier qui a choisi de participer aux tests PISA. Notons que les pays qui ne participent pas à l'enquête PISA (une centaine, tout de même) sont pour la plupart des pays en [voie de développement](https://fr.wikipedia.org/wiki/Pays_en_d%C3%A9veloppement), où les taux de scolarisation sont faibles. Si ceux-ci devaient être intégrés au test PISA, il serait probablement utile de considérer à minima un troisième cluster dont le score PISA moyen serait probablement plus faible que celle des deux clusters actuels.\n",
"\n",
"Enfin, au vu de ce qui précède, on est tenté de penser que le score PISA serait... \n",
"* Positivement corrélé avec : le fait d'être membre de l'OCDE, le PIB par tête, le fait d'investir dans l'éducation, l'espérance de vie à la naissance.\n",
"* Négativement corrélé avec : le nombre d'enfants par classe à l'école primaire.\n",
"\n",
"La partie suivante s'intéresse à l'existence et à la pertinence de ces corrélations."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### C. Étude de corrélation avec différents indicateurs\n",
"Au préalable, on décide de : \n",
"* Supprimer la Chine de notre base de données. En effet, comme indiqué précédemment : dans ce pays, les tests PISA n'ont été effectués que dans certaines régions particulièrement riches et très peu représentatives. \n",
"* Supprimer les lignes pour lesquelles des valeurs sont manquantes. Elles sont peu nombreuses et concernent principalement Taiwan et le Montenegro, pays pour lesquels la Banque Mondiale ne fournit pas tous les indicateurs économiques."
]
},
{
"cell_type": "code",
"execution_count": 47,
"metadata": {},
"outputs": [],
"source": [
"PISA_pivote_post2006 = PISA_pivote_post2006[PISA_pivote_post2006[\"Country\"] != \"China\"].dropna().copy()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"L'objectif de cette partie est d'étudier la corrélation entre le score moyen obtenu au score PISA et les indicateurs suivants :\n",
"* L'indicatrice d'appartenir ou non à l'OCDE (is_OECD)\n",
"* Les dépenses gouvernementales dans l'éducation en % du PIB (gov_exp)\n",
"* Le ratio nombre d'élèves / professeur à l'école primaire (pupil_teacher_ratio)\n",
"* L'espérance de vie à la naissance (life_expect)\n",
"* Le PIB par tête (GDP_per_capita)\n",
"\n",
"#### 1. Un premier aperçu général\n",
"Commençons par croiser le score PISA moyen obtenu par pays et par année avec ces différentes variables, une à une."
]
},
{
"cell_type": "code",
"execution_count": 48,
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Score moyen obtenu au test PISA en fonction de différents indicateurs. \n",
"Chaque point représente le score moyen d'un pays pour une année donnée.\n"
]
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"PISA_pivote_post2006['log_GDP_per_capita'] = np.log(PISA_pivote_post2006['GDP_per_capita'])\n",
"fig = px.scatter(PISA_pivote_post2006, x=\"log_GDP_per_capita\", y=\"Mean_score\", \n",
" hover_data=[\"Maths\", \"Reading\", \"Science\", \"is_OECD\"], hover_name=\"Country\", trendline=\"ols\",\n",
" title='Score PISA moyen en fonction du log(PIB par tête) ($ 2010), par pays et par année.')\n",
"fig.show()"
]
},
{
"cell_type": "code",
"execution_count": 50,
"metadata": {
"scrolled": false
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"
OLS Regression Results
\n",
"
\n",
"
Dep. Variable:
y
R-squared:
0.467
\n",
"
\n",
"
\n",
"
Model:
OLS
Adj. R-squared:
0.465
\n",
"
\n",
"
\n",
"
Method:
Least Squares
F-statistic:
255.4
\n",
"
\n",
"
\n",
"
Date:
Tue, 04 Feb 2020
Prob (F-statistic):
9.81e-42
\n",
"
\n",
"
\n",
"
Time:
15:19:30
Log-Likelihood:
-1471.9
\n",
"
\n",
"
\n",
"
No. Observations:
294
AIC:
2948.
\n",
"
\n",
"
\n",
"
Df Residuals:
292
BIC:
2955.
\n",
"
\n",
"
\n",
"
Df Model:
1
\n",
"
\n",
"
\n",
"
Covariance Type:
nonrobust
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
coef
std err
t
P>|t|
[0.025
0.975]
\n",
"
\n",
"
\n",
"
const
114.3908
22.147
5.165
0.000
70.804
157.978
\n",
"
\n",
"
\n",
"
x1
35.6109
2.229
15.980
0.000
31.225
39.997
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
Omnibus:
45.049
Durbin-Watson:
1.482
\n",
"
\n",
"
\n",
"
Prob(Omnibus):
0.000
Jarque-Bera (JB):
172.371
\n",
"
\n",
"
\n",
"
Skew:
-0.571
Prob(JB):
3.72e-38
\n",
"
\n",
"
\n",
"
Kurtosis:
6.573
Cond. No.
105.
\n",
"
\n",
"
Warnings: [1] Standard Errors assume that the covariance matrix of the errors is correctly specified."
],
"text/plain": [
"\n",
"\"\"\"\n",
" OLS Regression Results \n",
"==============================================================================\n",
"Dep. Variable: y R-squared: 0.467\n",
"Model: OLS Adj. R-squared: 0.465\n",
"Method: Least Squares F-statistic: 255.4\n",
"Date: Tue, 04 Feb 2020 Prob (F-statistic): 9.81e-42\n",
"Time: 15:19:30 Log-Likelihood: -1471.9\n",
"No. Observations: 294 AIC: 2948.\n",
"Df Residuals: 292 BIC: 2955.\n",
"Df Model: 1 \n",
"Covariance Type: nonrobust \n",
"==============================================================================\n",
" coef std err t P>|t| [0.025 0.975]\n",
"------------------------------------------------------------------------------\n",
"const 114.3908 22.147 5.165 0.000 70.804 157.978\n",
"x1 35.6109 2.229 15.980 0.000 31.225 39.997\n",
"==============================================================================\n",
"Omnibus: 45.049 Durbin-Watson: 1.482\n",
"Prob(Omnibus): 0.000 Jarque-Bera (JB): 172.371\n",
"Skew: -0.571 Prob(JB): 3.72e-38\n",
"Kurtosis: 6.573 Cond. No. 105.\n",
"==============================================================================\n",
"\n",
"Warnings:\n",
"[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
"\"\"\""
]
},
"execution_count": 50,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"results = px.get_trendline_results(fig)\n",
"results.px_fit_results[0].summary()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Interprétation :\n",
"* La p-valeur quasi-nulle du coefficient x1 permet de rejeter l'hypothèse nulle au seuil usuel de 5% et indique que le log(PIB par tête) a un effet statistiquement significatif sur le score PISA moyen.\n",
"* Étant donné qu'il s'agit d'un modèle en niveau-log, le coefficient x1 indique que : toutes choses égales par ailleurs, lorsque le PIB par tête d'un pays augmente de 1%, on estime que son score PISA moyen va augmenter de 36 points en moyenne.\n",
"\n",
"Le passage en logarithme permet donc bien d'obtenir un effet explicatif satisfaisant pour la variable PIB par tête.\n",
"\n",
"#### 3. Régression multiple sur les 5 variables d'intérêts\n",
"\n",
"On souhaite désormais estimer le modèle suivant :\n",
"$Mean\\_score = const + x_1 * log\\_GDP\\_per\\_capita + x_2 * gov\\_exp + x_3 * life\\_expect + x_4 * pupil\\_teacher\\_ratio + x_5 * is\\_OECD$"
]
},
{
"cell_type": "code",
"execution_count": 51,
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"/Users/pierre/anaconda3/lib/python3.7/site-packages/numpy/core/fromnumeric.py:2389: FutureWarning:\n",
"\n",
"Method .ptp is deprecated and will be removed in a future version. Use numpy.ptp instead.\n",
"\n"
]
},
{
"data": {
"text/html": [
"
\n",
"
OLS Regression Results
\n",
"
\n",
"
Dep. Variable:
Mean_score
R-squared:
0.527
\n",
"
\n",
"
\n",
"
Model:
OLS
Adj. R-squared:
0.519
\n",
"
\n",
"
\n",
"
Method:
Least Squares
F-statistic:
64.22
\n",
"
\n",
"
\n",
"
Date:
Tue, 04 Feb 2020
Prob (F-statistic):
7.27e-45
\n",
"
\n",
"
\n",
"
Time:
15:19:30
Log-Likelihood:
-1454.1
\n",
"
\n",
"
\n",
"
No. Observations:
294
AIC:
2920.
\n",
"
\n",
"
\n",
"
Df Residuals:
288
BIC:
2942.
\n",
"
\n",
"
\n",
"
Df Model:
5
\n",
"
\n",
"
\n",
"
Covariance Type:
nonrobust
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
coef
std err
t
P>|t|
[0.025
0.975]
\n",
"
\n",
"
\n",
"
const
113.6410
50.927
2.231
0.026
13.404
213.878
\n",
"
\n",
"
\n",
"
log_GDP_per_capita
19.9852
4.024
4.967
0.000
12.065
27.905
\n",
"
\n",
"
\n",
"
gov_exp
0.2412
1.699
0.142
0.887
-3.103
3.585
\n",
"
\n",
"
\n",
"
life_expect
1.8157
0.892
2.035
0.043
0.059
3.572
\n",
"
\n",
"
\n",
"
pupil_teacher_ratio
-0.2575
0.515
-0.500
0.618
-1.272
0.757
\n",
"
\n",
"
\n",
"
is_OECD
28.9419
5.384
5.375
0.000
18.345
39.539
\n",
"
\n",
"
\n",
"
\n",
"
\n",
"
Omnibus:
16.176
Durbin-Watson:
2.097
\n",
"
\n",
"
\n",
"
Prob(Omnibus):
0.000
Jarque-Bera (JB):
42.988
\n",
"
\n",
"
\n",
"
Skew:
0.083
Prob(JB):
4.63e-10
\n",
"
\n",
"
\n",
"
Kurtosis:
4.866
Cond. No.
2.04e+03
\n",
"
\n",
"
Warnings: [1] Standard Errors assume that the covariance matrix of the errors is correctly specified. [2] The condition number is large, 2.04e+03. This might indicate that there are strong multicollinearity or other numerical problems."
],
"text/plain": [
"\n",
"\"\"\"\n",
" OLS Regression Results \n",
"==============================================================================\n",
"Dep. Variable: Mean_score R-squared: 0.527\n",
"Model: OLS Adj. R-squared: 0.519\n",
"Method: Least Squares F-statistic: 64.22\n",
"Date: Tue, 04 Feb 2020 Prob (F-statistic): 7.27e-45\n",
"Time: 15:19:30 Log-Likelihood: -1454.1\n",
"No. Observations: 294 AIC: 2920.\n",
"Df Residuals: 288 BIC: 2942.\n",
"Df Model: 5 \n",
"Covariance Type: nonrobust \n",
"=======================================================================================\n",
" coef std err t P>|t| [0.025 0.975]\n",
"---------------------------------------------------------------------------------------\n",
"const 113.6410 50.927 2.231 0.026 13.404 213.878\n",
"log_GDP_per_capita 19.9852 4.024 4.967 0.000 12.065 27.905\n",
"gov_exp 0.2412 1.699 0.142 0.887 -3.103 3.585\n",
"life_expect 1.8157 0.892 2.035 0.043 0.059 3.572\n",
"pupil_teacher_ratio -0.2575 0.515 -0.500 0.618 -1.272 0.757\n",
"is_OECD 28.9419 5.384 5.375 0.000 18.345 39.539\n",
"==============================================================================\n",
"Omnibus: 16.176 Durbin-Watson: 2.097\n",
"Prob(Omnibus): 0.000 Jarque-Bera (JB): 42.988\n",
"Skew: 0.083 Prob(JB): 4.63e-10\n",
"Kurtosis: 4.866 Cond. No. 2.04e+03\n",
"==============================================================================\n",
"\n",
"Warnings:\n",
"[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n",
"[2] The condition number is large, 2.04e+03. This might indicate that there are\n",
"strong multicollinearity or other numerical problems.\n",
"\"\"\""
]
},
"execution_count": 51,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"X = PISA_pivote_post2006[['log_GDP_per_capita', 'gov_exp', 'life_expect', 'pupil_teacher_ratio', 'is_OECD']]\n",
"X = sm.add_constant(X)\n",
"y = PISA_pivote_post2006['Mean_score']\n",
"\n",
"model = sm.OLS(y, X)\n",
"results = model.fit()\n",
"results.summary()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"**Étude des p-valeurs** :\n",
"* Les p-valeurs associées aux variables gov_exp et pupil_teacher_ratio sont énormes et ne permettent pas de rejeter l'hypothèse nulle. Ces variables n'ont donc aucun effet statistiquement significatif sur le score PISA moyen, contrairement à ce qu'on aurait pu intuiter.\n",
"* Les p-valeurs associées aux variables log_GDP_per_capita, life_expect et is_OECD sont quasi-nulles ce qui permet de rejeter l'hypothèse nulle à tous les seuils usuels.\n",
"\n",
"**Étude des coefficients significatifs** : toutes choses égales par ailleurs...\n",
"* Le coefficient de log_GDP_per_capita indique que, lorsque le PIB par tête augmente d'1%, on prédit que le score PISA moyen va augmenter de 20 points.\n",
"* Le coefficient de life_expect indique que, lorsque l'espérance de vie à la naissance augmente d'1 an, on prédit que le score PISA moyen va augmenter de 2 points environ.\n",
"* Le coefficient de is_OECD indique que, lorsqu'un pays est membre de l'OCDE, on prédit que son score PISA moyen sera 25 points supérieur à celui d'un pays non-membre de l'OCDE.\n",
"\n",
"**Commentaire général sur la pertinence** de cette régression : malgré son potentiel prédictif intéressant, celle-ci ne permet absolument pas d'identifier un effet causal en raison de différents biais.\n",
"* Un biais de sélection : les pays sélectionnés ne sont pas représentatifs de l'ensemble des pays du monde puisque les pays les moins développés ne sont pas représentés.\n",
"* Un biais de variable omise : les régresseurs sont probablement corrélés au terme d'erreur. Un exemple, parmi bien d'autres : notre modèle n'inclut pas l'âge jusqu'auquel la scolarisation est obligatoire. Or, on peut penser que cet indicateur serait positivement corrélé au score PISA et aux dépenses du gouvernement dans l'éducation."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### D. Prédiction du score moyen pour les pays n'ayant pas participé au test PISA\n",
"En 2018, seuls 78 pays ont participé au test PISA. Ils sont donc une centaine à ne pas y avoir pris part. À partir de notre modèle ci-dessus, et malgré ses imperfections notables, on va essayer de prédire quel aurait été le score des pays qui n'ont pas pris part au test, pour chaque édition. Ré-utilisons les données scrappées dans la partie 1 afin de construire une table contenant les variables explicatives pour tous les pays du monde, y compris ceux qui n'ont pas participé."
]
},
{
"cell_type": "code",
"execution_count": 52,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"fig = px.scatter(PISA_predict, x=\"Mean_score\", y=\"Mean_score_predict\", hover_name=\"Country\", \n",
" hover_data=[\"Year\", \"GDP_per_capita\", \"life_expect\", \"gov_exp\", \"pupil_teacher_ratio\"], \n",
" color=\"Prediction_error\", animation_frame=\"Year\")\n",
"bissectrice = np.linspace(300,600)\n",
"fig.add_trace(go.Scatter(x=bissectrice, y=bissectrice, mode='lines', name='lines'))\n",
"fig.show()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"* Les pays en dessous de la bissectrice obtiennent un score moins bon que celui que l'on aurait prédit.\n",
"* Les pays au dessus de la bissectrice obtiennent un score meilleur que celui que l'on aurait prédit.\n",
"\n",
"Certains cas s'expliquent assez bien avec l'intuition. Par exemple, en 2018, le Qatar a un score bien plus faible que celui anticipé (quasiment 70 pts d'erreur). Notre prédiction est totalement faussée par un PIB / tête très élevé (63k €) qui ne rend pas compte des inégalités socio-économiques dans ce pays, par exemple."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"#### 3. Que retenir ?\n",
"Notre modèle est donc très imprécis et les résultats obtenus sont donc à considérer avec vigilance. On pouvait s'y attendre vu le R2 obtenu ci-dessus (51%). Une solution simple consisterait à augmenter le nombre de variables utilisé, tout en se laissant guider par l'intuition pour leur choix, en observant notamment les pays pour lesquels l'écart est trop faible et en essayant de trouver des indicateurs qui pourraient expliquer ces écarts (ex : mesure des inégalités, etc.). La difficulté toutefois, à laquelle nous avons dû faire face, consiste à trouver des indicateurs faciles d'accès et qui soient disponibles pour tous les pays."
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"## Conclusion générale\n",
"À l'aide de **données scrappées** sur divers sites (Wikipedia, Banque Mondiale, CIA...), nous avons pu analyser les résultats de l'enquête PISA depuis sa création en 2000, en observant l'**évolution de leur répartition au fil du temps**, et en effectuant différents focus sur le cas de la **France** et de l'**OCDE**. Pour revenir sur le cas de la France, on a vu qu'il était possible de **\"faire parler\"** les chiffres de diverses manières : dans l'absolu, les performances baissent, mais en relatif cette tendance est conforme à celle de l'OCDE. Néanmoins, la question se pose de savoir s'il est toujours pertinent de se comparer aux pays de l'OCDE, un groupe dont la cohérence est discutable, qui comprend des pays comme le Chili ou le Mexique, dont la situation économique est très différente de celle de la France, et qui exclut en même temps des pays riches, dont la situation économique est plus proche de celle de la France, comme Singapour ou la Corée du Sud.\n",
"\n",
"Dans la troisième partie, on a constaté que certains pays affichaient des performances très différentes en mathématiques / lecture / sciences et qu'il était donc utile de **ne pas se limiter à la simple comparaison de scores moyens**. On a ensuite pu observer qu'il était possible et assez naturel de diviser les pays participants en **deux clusters à la performance homogène**. On a remarqué que les pays composant le cluster le plus performant avaient des points communs : ils sont en moyenne beaucoup plus riches, ont une espérance de vie supérieure à la moyenne, ils investissent plus dans l'éducation, ils ont un ratio élèves / professeur à l'école primaire inférieur à la moyenne, et ils sont majoritairement membres de l'OCDE. Partant de ce résultat, on a souhaité approfondir la corrélation avec ces indicateurs en proposant un modèle fondé sur une **régression multivariée**. Enfin, à l'aide de ce modèle, on a proposé une **prédiction du score PISA** de tous les pays du monde, y compris tous ceux qui n'y ont jamais participé, par année. On a néanmoins été contraints d'observer que notre modèle était très imprécis, ce qui nous a poussé à émettre différentes critiques à son sujet, et à faire plusieurs suggestions permettant de l'améliorer."
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.7.3"
}
},
"nbformat": 4,
"nbformat_minor": 2
}