University graduates employment survey.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Employment rate and salary trends between 2013 - 2017 for Accountancy and Business Graduates"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "toc": true
   },
   "source": [
    "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Quick-Summary:\" data-toc-modified-id=\"Quick-Summary:-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>Quick Summary:</a></span></li><li><span><a href=\"#Load-all-relevant-data\" data-toc-modified-id=\"Load-all-relevant-data-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>Load all relevant data</a></span><ul class=\"toc-item\"><li><span><a href=\"#Load-dataset-and-metadata\" data-toc-modified-id=\"Load-dataset-and-metadata-2.1\"><span class=\"toc-item-num\">2.1&nbsp;&nbsp;</span>Load dataset and metadata</a></span></li></ul></li><li><span><a href=\"#Data-preprocessing\" data-toc-modified-id=\"Data-preprocessing-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>Data preprocessing</a></span></li><li><span><a href=\"#The-reason-why-I-choose-not-to-compare-between-universities-is-due-to-a-couple-of-issues.\" data-toc-modified-id=\"The-reason-why-I-choose-not-to-compare-between-universities-is-due-to-a-couple-of-issues.-4\"><span class=\"toc-item-num\">4&nbsp;&nbsp;</span>The reason why I choose not to compare between universities is due to a couple of issues.</a></span><ul class=\"toc-item\"><li><span><a href=\"#Proof-as-to-why-we-cannot-simply-take-the-mean-salary-of-hons-and-no-hons-to-be-the-average-of-both\" data-toc-modified-id=\"Proof-as-to-why-we-cannot-simply-take-the-mean-salary-of-hons-and-no-hons-to-be-the-average-of-both-4.1\"><span class=\"toc-item-num\">4.1&nbsp;&nbsp;</span>Proof as to why we cannot simply take the mean salary of hons and no hons to be the average of both</a></span></li></ul></li><li><span><a href=\"#Insights\" data-toc-modified-id=\"Insights-5\"><span class=\"toc-item-num\">5&nbsp;&nbsp;</span>Insights</a></span></li></ul></div>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Quick Summary:\n",
    "\n",
    "Accountancy degree is a much safer option in terms of employability as compared to a Business degree (Although the employability of each major is unknown). Only about 5% ~ 10% of Accountancy graduates run risk of facing unemployment within the first 6 months of finishing their final examinations as compared to about 15% ~ 20% of the Business graduates.\n",
    "\n",
    "However, Business graduates generally earn more. Hence, if one's goal is to start with a relatively higher pay, getting a Business degree is recommended since the pay for Accountancy fresh graduates have been rather stagnant.\n",
    "\n",
    "Out of the three universities, NUS Business degree has the highest median salary as compared to its counterparts. Unfortunately, the salaries for the major within each Business degree is unknown."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load all relevant data\n",
    "\n",
    "Data source: https://data.gov.sg/dataset/graduate-employment-survey-ntu-nus-sit-smu-sutd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load dataset and metadata"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": false
   },
   "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>university</th>\n",
       "      <th>school</th>\n",
       "      <th>degree</th>\n",
       "      <th>employment_rate_overall</th>\n",
       "      <th>employment_rate_ft_perm</th>\n",
       "      <th>basic_monthly_mean</th>\n",
       "      <th>basic_monthly_median</th>\n",
       "      <th>gross_monthly_mean</th>\n",
       "      <th>gross_monthly_median</th>\n",
       "      <th>gross_mthly_25_percentile</th>\n",
       "      <th>gross_mthly_75_percentile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2013</td>\n",
       "      <td>Nanyang Technological University</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Accountancy and Business</td>\n",
       "      <td>97.4</td>\n",
       "      <td>96.1</td>\n",
       "      <td>3701</td>\n",
       "      <td>3200</td>\n",
       "      <td>3727</td>\n",
       "      <td>3350</td>\n",
       "      <td>2900</td>\n",
       "      <td>4000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2013</td>\n",
       "      <td>Nanyang Technological University</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Accountancy (3-yr direct Honours Programme)</td>\n",
       "      <td>97.1</td>\n",
       "      <td>95.7</td>\n",
       "      <td>2850</td>\n",
       "      <td>2700</td>\n",
       "      <td>2938</td>\n",
       "      <td>2700</td>\n",
       "      <td>2700</td>\n",
       "      <td>2900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2013</td>\n",
       "      <td>Nanyang Technological University</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Business (3-yr direct Honours Programme)</td>\n",
       "      <td>90.9</td>\n",
       "      <td>85.7</td>\n",
       "      <td>3053</td>\n",
       "      <td>3000</td>\n",
       "      <td>3214</td>\n",
       "      <td>3000</td>\n",
       "      <td>2700</td>\n",
       "      <td>3500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2013</td>\n",
       "      <td>Nanyang Technological University</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Business and Computing</td>\n",
       "      <td>87.5</td>\n",
       "      <td>87.5</td>\n",
       "      <td>3557</td>\n",
       "      <td>3400</td>\n",
       "      <td>3615</td>\n",
       "      <td>3400</td>\n",
       "      <td>3000</td>\n",
       "      <td>4100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2013</td>\n",
       "      <td>Nanyang Technological University</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Aerospace Engineering</td>\n",
       "      <td>95.3</td>\n",
       "      <td>95.3</td>\n",
       "      <td>3494</td>\n",
       "      <td>3500</td>\n",
       "      <td>3536</td>\n",
       "      <td>3500</td>\n",
       "      <td>3100</td>\n",
       "      <td>3816</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year                        university  \\\n",
       "0  2013  Nanyang Technological University   \n",
       "1  2013  Nanyang Technological University   \n",
       "2  2013  Nanyang Technological University   \n",
       "3  2013  Nanyang Technological University   \n",
       "4  2013  Nanyang Technological University   \n",
       "\n",
       "                                          school  \\\n",
       "0  College of Business (Nanyang Business School)   \n",
       "1  College of Business (Nanyang Business School)   \n",
       "2  College of Business (Nanyang Business School)   \n",
       "3  College of Business (Nanyang Business School)   \n",
       "4                         College of Engineering   \n",
       "\n",
       "                                        degree employment_rate_overall  \\\n",
       "0                     Accountancy and Business                    97.4   \n",
       "1  Accountancy (3-yr direct Honours Programme)                    97.1   \n",
       "2     Business (3-yr direct Honours Programme)                    90.9   \n",
       "3                       Business and Computing                    87.5   \n",
       "4                        Aerospace Engineering                    95.3   \n",
       "\n",
       "  employment_rate_ft_perm basic_monthly_mean basic_monthly_median  \\\n",
       "0                    96.1               3701                 3200   \n",
       "1                    95.7               2850                 2700   \n",
       "2                    85.7               3053                 3000   \n",
       "3                    87.5               3557                 3400   \n",
       "4                    95.3               3494                 3500   \n",
       "\n",
       "  gross_monthly_mean gross_monthly_median gross_mthly_25_percentile  \\\n",
       "0               3727                 3350                      2900   \n",
       "1               2938                 2700                      2700   \n",
       "2               3214                 3000                      2700   \n",
       "3               3615                 3400                      3000   \n",
       "4               3536                 3500                      3100   \n",
       "\n",
       "  gross_mthly_75_percentile  \n",
       "0                      4000  \n",
       "1                      2900  \n",
       "2                      3500  \n",
       "3                      4100  \n",
       "4                      3816  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Enter your own path if necessary\n",
    "df = pd.read_csv('survey.csv', sep = ';')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "# Metadata for Graduate Employment Survey - NTU, NUS, SIT, SMU & SUTD\n",
      "---\n",
      "Identifier: 'aa812691-df21-4c0a-b553-85336695f872'\n",
      "Name: 'graduate-employment-survey-ntu-nus-sit-smu-sutd'\n",
      "Title: 'Graduate Employment Survey - NTU, NUS, SIT, SMU & SUTD'\n",
      "Description:\n",
      "  - 'The Graduate Employment Survey (GES) is jointly conducted by NTU, NUS, SMU, SIT\n",
      "    (from 2014) and SUTD (from 2015) annually to survey the employment conditions\n",
      "    of graduates about six months after their final examinations.  The Ministry of\n",
      "    Education (MOE) publishes the results of key employment indicators of the survey\n",
      "    to provide prospective students with timely and comparable data to assist them\n",
      "    in making informed course decisions.  Graduates from NTU, NUS and SMU were surveyed\n",
      "    in Nov.  Due to their different academic calendars, graduates from SUTD and SIT\n",
      "    were surveyed in Feb and Mar.'\n",
      "  - ''\n",
      "  - '2017 GES:'\n",
      "  - ''\n",
      "  - '*  6,042 full-time graduates and 30 follow-up graduates from NTU were surveyed\n",
      "    and the overall response rate obtained were 84.3% and 100% respectively.'\n",
      "  - '*  6,466 full-time graduates and 825 follow-up from NUS were surveyed and the\n",
      "    overall response rate obtained was 80.6% and 72.1% respectively.'\n",
      "  - '*  1,779 full-time graduates and 114 follow-up graduates from SMU were surveyed\n",
      "    and the overall response rate obtained was 77.6% and 75.4% respectively.'\n",
      "  - '*  228 graduates from SUTD were surveyed and the overall response rate obtained\n",
      "    was 85.4%.'\n",
      "  - '*  1,530 graduates from SIT were surveyed and the overall response rate obtained\n",
      "    was 85.6%.'\n",
      "  - ''\n",
      "  - '2016 GES:'\n",
      "  - ''\n",
      "  - '*  5,836 graduates from NTU were surveyed and the overall response rate obtained\n",
      "    was 77.0%.'\n",
      "  - '*  6,314 graduates from NUS were surveyed and the overall response rate obtained\n",
      "    was 80.4%.'\n",
      "  - '*  1,803 graduates from SMU were surveyed and the overall response rate obtained\n",
      "    was 74.1%.'\n",
      "  - '*  246 graduates from SUTD were surveyed and the overall response rate obtained\n",
      "    was 83.7%.'\n",
      "  - '*  1,230 graduates from SIT were surveyed and the overall response rate obtained\n",
      "    was 84.6%.'\n",
      "  - ''\n",
      "  - '2015 GES:'\n",
      "  - ''\n",
      "  - '*  5,774 graduates from NTU were surveyed and the overall response rate obtained\n",
      "    was 71.9%.'\n",
      "  - '*  6,188 graduates from NUS were surveyed and the overall response rate obtained\n",
      "    was 74.8%.'\n",
      "  - '* 1,638 graduates from SMU were surveyed and the overall response rate obtained\n",
      "    was 76.2%.'\n",
      "  - '* 298 graduates from SUTD were surveyed and the overall response rate was 86.2%.'\n",
      "  - '* 1,363 graduates from SIT were surveyed and the overall response rate obtained\n",
      "    was 78.9%.'\n",
      "  - ''\n",
      "  - '2014 GES:  14,868 graduates from NTU, NUS, SIT and SMU were surveyed and the\n",
      "    overall response rate obtained was 75%.'\n",
      "  - ''\n",
      "  - '2013 GES:  15,365 graduates from NTU, NUS and SMU were surveyed and the overall\n",
      "    response rate obtained was 72%.'\n",
      "  - ''\n",
      "  - 'Notes:'\n",
      "  - ''\n",
      "  - '\\*  SMU''s courses are direct 4-yr programmes and graduates could be awarded\n",
      "    Cum Laude and above, Merit, High Merit or no awards.  The data for SMU''s courses\n",
      "    is displayed in two categories: (i) overall results for all graduates within the\n",
      "    course regardless of the award they attained, and (ii) results for the graduates\n",
      "    awarded Cum Laude and above.  Should the graduates undertake more than 1 Degree\n",
      "    (e.g. Dual Degree programme), they have been classified based on their first Degree.'\n",
      "  - ''\n",
      "  - '^  Data on Chinese Medicine graduates is obtained from a follow-up survey on\n",
      "    those who graduated 1 year ago (e.g. for 2017 GES survey, data is obtained from\n",
      "    2016 graduates) after they have completed their practical training.'\n",
      "  - ''\n",
      "  - '\\**  No data is shown due to the small number of graduates or low response rates.'\n",
      "  - ''\n",
      "  - '\\##  Data on law, medical and pharmacy graduates are obtained from a follow-up\n",
      "    survey on those who graduated 1 year ago (e.g. for 2017 GES survey, data is obtained\n",
      "    from 2016 graduates) after they have completed their 1-year practical law course/pupillage/housemanship/first-year\n",
      "    residency/practical training.  Data on architecture graduates are obtained from\n",
      "    a follow-up survey on those who graduated 3 years ago (e.g. for 2017 GES survey,\n",
      "    data is obtained from 2014 graduates) after they have completed their practical\n",
      "    training.'\n",
      "  - ''\n",
      "  - '^^  Data includes the employment figures of graduates from Bachelor of Engineering\n",
      "    (Bioengineering), which was merged into Bachelor of Engineering (Biomedical Engineering)\n",
      "    in 2014.'\n",
      "  - ''\n",
      "  - '++  Data on law graduates are obtained from a follow-up survey on those who graduated\n",
      "    1 year ago (e.g. for 2017 GES survey, data is obtained from 2016 graduates) after\n",
      "    they have completed their practical law course/pupillage.'\n",
      "  - ''\n",
      "  - '\\###  Data on SUTD''s Bachelor of Science (Architecture and Sustainable Design)\n",
      "    graduates will be obtained from a follow-up survey on the graduates after they\n",
      "    have completed their practical training.'\n",
      "  - ''\n",
      "  - '^^^  Data is based on a sample size of fewer than 30 respondents.'\n",
      "  - ''\n",
      "  - ''\n",
      "  - ''\n",
      "  - ''\n",
      "  - '**Disclaimer:**'\n",
      "  - 'The data published is provided on an “as is” basis without any warranties of\n",
      "    any kind.  To the fullest extent permitted by law, MOE does not warrant and hereby\n",
      "    disclaims any warranty as to the accuracy, correctness, completeness, reliability\n",
      "    or fitness for any particular purpose of such data.  The reader assumes all responsibility\n",
      "    and risk for the use of such data.  MOE shall not be liable for any damage or\n",
      "    loss of any kind caused as a result (directly or indirectly) of the use of such\n",
      "    data, including but not limited to any damage or loss suffered as a result of\n",
      "    reliance on the data published.'\n",
      "Topics:\n",
      "  - 'Education'\n",
      "Keywords:\n",
      "  - 'Degree'\n",
      "  - 'Employment'\n",
      "  - 'Jobs'\n",
      "  - 'Salaries'\n",
      "  - 'University'\n",
      "Publisher:\n",
      "  Name: 'Ministry of Education'\n",
      "  Admin 1:\n",
      "    Name: 'Christina Leow'\n",
      "    Department: 'Research & Management Information Division'\n",
      "    Email: 'christina_leow@moe.gov.sg'\n",
      "  Admin 2:\n",
      "    Name: 'Chng Lung Huang'\n",
      "    Department: 'Research & Management Information Division'\n",
      "    Email: 'chng_lung_huang@moe.gov.sg'\n",
      "Sources:\n",
      "  - 'Ministry of Education'\n",
      "License: 'https://data.gov.sg/open-data-licence'\n",
      "Frequency: 'Annual'\n",
      "Coverage: '2013-01-01 to 2017-12-31'\n",
      "Last Updated: '2019-01-24T10:30:48.808083'\n",
      "Resources:\n",
      "  -\n",
      "    Identifier: '9326ca53-9153-4a9c-b93f-8ae032637b70'\n",
      "    Title: 'Graduate Employment Survey - NTU, NUS, SIT, SMU & SUTD'\n",
      "    Url: 'https://storage.data.gov.sg/graduate-employment-survey-ntu-nus-sit-smu-sutd/resources/graduate-employment-survey-ntu-nus-sit-smu-sutd-2019-01-23T07-49-57Z.csv'\n",
      "    Format: 'CSV'\n",
      "    Coverage: '2013-01-01 to 2017-12-31'\n",
      "    Last Updated: '2019-01-23T07:49:57.874203'\n",
      "    Schema:\n",
      "      -\n",
      "        Name: 'year'\n",
      "        Title: 'Year'\n",
      "        Type: 'datetime'\n",
      "        Sub Type: 'year'\n",
      "        Format: 'YYYY'\n",
      "      -\n",
      "        Name: 'university'\n",
      "        Title: 'University'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "      -\n",
      "        Name: 'school'\n",
      "        Title: 'School'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "        Description: '\"na\" : Data not available or not applicable'\n",
      "      -\n",
      "        Name: 'degree'\n",
      "        Title: 'Degree'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "      -\n",
      "        Name: 'employment_rate_overall'\n",
      "        Title: 'Overall Employment Rate (%)'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "        Description:\n",
      "          - 'Overall employment rate refers to the number of graduates working in\n",
      "            full-time permanent, part-time, temporary or freelance employment, as\n",
      "            a proportion of graduates in the labour force (i.e. those who were working,\n",
      "            or not working but actively looking and available for work) approximately\n",
      "            6 months after completing their final examinations.'\n",
      "          - '\"na\" : Data not available or not applicable'\n",
      "      -\n",
      "        Name: 'employment_rate_ft_perm'\n",
      "        Title: 'Full-Time Permanent Employment Rate (%)'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "        Description:\n",
      "          - 'Full-time permanent employment rate refers to the number of graduates\n",
      "            in employment of at least 35 hours a week and where the employment is\n",
      "            not temporary (including contracts of one year or more), as a proportion\n",
      "            of graduates in the labour force (i.e. those who were working, or not\n",
      "            working but actively looking and available for work) approximately 6 months\n",
      "            after completing their final examinations.'\n",
      "          - '\"na\" : Data not available or not applicable'\n",
      "      -\n",
      "        Name: 'basic_monthly_mean'\n",
      "        Title: 'Basic Monthly Salary - Mean (S$)'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "        Description:\n",
      "          - 'Basic monthly salary pertains only to full-time permanently employed\n",
      "            graduates. It comprises basic pay before deduction of the employee’s CPF\n",
      "            contributions and personal income tax. Employer’s CPF contributions, bonuses,\n",
      "            stock options, overtime payments, commissions, allowances, other regular\n",
      "            cash payments, lump sum payments, and payments-in-kind are excluded.'\n",
      "          - '\"na\" : Data not available or not applicable'\n",
      "      -\n",
      "        Name: 'basic_monthly_median'\n",
      "        Title: 'Basic Monthly Salary - Median (S$)'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "        Description:\n",
      "          - 'Basic monthly salary pertains only to full-time permanently employed\n",
      "            graduates. It comprises basic pay before deduction of the employee’s CPF\n",
      "            contributions and personal income tax. Employer’s CPF contributions, bonuses,\n",
      "            stock options, overtime payments, commissions, allowances, other regular\n",
      "            cash payments, lump sum payments, and payments-in-kind are excluded.'\n",
      "          - '\"na\" : Data not available or not applicable'\n",
      "      -\n",
      "        Name: 'gross_monthly_mean'\n",
      "        Title: 'Gross Monthly Salary - Mean (S$)'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "        Description:\n",
      "          - 'Gross monthly salary pertains only to full-time permanently employed\n",
      "            graduates. It comprises basic salary, overtime payments, commissions,\n",
      "            fixed allowances, and other regular cash payments, before deduction of\n",
      "            the employee’s CPF contributions and personal income tax. Employer’s CPF\n",
      "            contributions, bonuses, stock options, lump sum payments, and payments-in-kind\n",
      "            are excluded.'\n",
      "          - '\"na\" : Data not available or not applicable'\n",
      "      -\n",
      "        Name: 'gross_monthly_median'\n",
      "        Title: 'Gross Monthly Salary - Median (S$)'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "        Description:\n",
      "          - 'Gross monthly salary pertains only to full-time permanently employed\n",
      "            graduates. It comprises basic salary, overtime payments, commissions,\n",
      "            fixed allowances, and other regular cash payments, before deduction of\n",
      "            the employee’s CPF contributions and personal income tax. Employer’s CPF\n",
      "            contributions, bonuses, stock options, lump sum payments, and payments-in-kind\n",
      "            are excluded.'\n",
      "          - '\"na\" : Data not available or not applicable'\n",
      "      -\n",
      "        Name: 'gross_mthly_25_percentile'\n",
      "        Title: 'Gross Monthly Salary - 25th Percentile (S$)'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "        Description:\n",
      "          - 'Gross monthly salary pertains only to full-time permanently employed\n",
      "            graduates. It comprises basic salary, overtime payments, commissions,\n",
      "            fixed allowances, and other regular cash payments, before deduction of\n",
      "            the employee’s CPF contributions and personal income tax. Employer’s CPF\n",
      "            contributions, bonuses, stock options, lump sum payments, and payments-in-kind\n",
      "            are excluded.'\n",
      "          - '\"na\" : Data not available or not applicable'\n",
      "      -\n",
      "        Name: 'gross_mthly_75_percentile'\n",
      "        Title: 'Gross Monthly Salary - 75th Percentile (S$)'\n",
      "        Type: 'text'\n",
      "        Sub Type: 'general'\n",
      "        Description:\n",
      "          - 'Gross monthly salary pertains only to full-time permanently employed\n",
      "            graduates. It comprises basic salary, overtime payments, commissions,\n",
      "            fixed allowances, and other regular cash payments, before deduction of\n",
      "            the employee’s CPF contributions and personal income tax. Employer’s CPF\n",
      "            contributions, bonuses, stock options, lump sum payments, and payments-in-kind\n",
      "            are excluded.'\n",
      "          - '\"na\" : Data not available or not applicable'\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Enter your own path where you saved your metadata-survey.txt\n",
    "f = open(\"metadata-survey.txt\", \"r\", encoding = 'UTF-8')\n",
    "print(f.read())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data preprocessing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['Nanyang Technological University',\n",
       "       'National University of Singapore',\n",
       "       'Singapore Management University',\n",
       "       'Singapore Institute of Technology',\n",
       "       'Singapore University of Technology and Design'], dtype=object)"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check unique university names to ensure a given university is not spelled differently\n",
    "df['university'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['NTU', 'NUS', 'SMU', 'SIT', 'SUTD'], dtype=object)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Replace university names to their respective acronyms\n",
    "df['university'] = df['university'].replace(['Nanyang Technological University',\n",
    "       'National University of Singapore',\n",
    "       'Singapore Management University',\n",
    "       'Singapore Institute of Technology',\n",
    "       'Singapore University of Technology and Design'], ['NTU', 'NUS', 'SMU', 'SIT', 'SUTD'])\n",
    "\n",
    "# Check to ensure the result \n",
    "df['university'].unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 570 entries, 0 to 569\n",
      "Data columns (total 12 columns):\n",
      "year                         570 non-null int64\n",
      "university                   570 non-null object\n",
      "school                       570 non-null object\n",
      "degree                       570 non-null object\n",
      "employment_rate_overall      570 non-null object\n",
      "employment_rate_ft_perm      570 non-null object\n",
      "basic_monthly_mean           570 non-null object\n",
      "basic_monthly_median         570 non-null object\n",
      "gross_monthly_mean           570 non-null object\n",
      "gross_monthly_median         570 non-null object\n",
      "gross_mthly_25_percentile    570 non-null object\n",
      "gross_mthly_75_percentile    570 non-null object\n",
      "dtypes: int64(1), object(11)\n",
      "memory usage: 53.5+ KB\n",
      "None\n",
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 570 entries, 0 to 569\n",
      "Data columns (total 12 columns):\n",
      "year                         570 non-null int64\n",
      "university                   570 non-null object\n",
      "school                       558 non-null object\n",
      "degree                       570 non-null object\n",
      "employment_rate_overall      513 non-null object\n",
      "employment_rate_ft_perm      513 non-null object\n",
      "basic_monthly_mean           513 non-null object\n",
      "basic_monthly_median         513 non-null object\n",
      "gross_monthly_mean           513 non-null object\n",
      "gross_monthly_median         513 non-null object\n",
      "gross_mthly_25_percentile    513 non-null object\n",
      "gross_mthly_75_percentile    513 non-null object\n",
      "dtypes: int64(1), object(11)\n",
      "memory usage: 53.5+ KB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "# While it may seem like there are no cases of NA. That might be misleading as NA is currently recognised as string.\n",
    "# Looking at the metedata, it is stated that data that are not available or not applicable are stored as 'na'\n",
    "# This means that we can get python to identify 'na' as NA before deciding how to treat it\n",
    "print(df.info())\n",
    "\n",
    "# Convert 'na' to NaN\n",
    "df = df.replace('na', np.nan)\n",
    "\n",
    "# Notice the change in the number of non-null object. This means that na is now recognised\n",
    "print(df.info())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "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>university</th>\n",
       "      <th>school</th>\n",
       "      <th>degree</th>\n",
       "      <th>employment_rate_overall</th>\n",
       "      <th>employment_rate_ft_perm</th>\n",
       "      <th>basic_monthly_mean</th>\n",
       "      <th>basic_monthly_median</th>\n",
       "      <th>gross_monthly_mean</th>\n",
       "      <th>gross_monthly_median</th>\n",
       "      <th>gross_mthly_25_percentile</th>\n",
       "      <th>gross_mthly_75_percentile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>2013</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Sciences</td>\n",
       "      <td>Biomedical Sciences **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>2013</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Sciences</td>\n",
       "      <td>Mathematics &amp; Economics **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Computational Biology) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Computer Engineering) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Design &amp; Environment</td>\n",
       "      <td>Bachelor of Arts (Architecture) ** ##</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>66</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Yong Siew Toh Conservatory of Music</td>\n",
       "      <td>Bachelor of Music **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>67</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Applied Science **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Science (Computational Biology) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>2014</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Sciences</td>\n",
       "      <td>Biomedical Science **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>127</th>\n",
       "      <td>2014</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Computational Biology) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>128</th>\n",
       "      <td>2014</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Computer Engineering) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>151</th>\n",
       "      <td>2014</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Yong Siew Toh Conservatory of Music</td>\n",
       "      <td>Bachelor of Music **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>152</th>\n",
       "      <td>2014</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Applied Science **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>156</th>\n",
       "      <td>2014</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Science (Computational Biology) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>199</th>\n",
       "      <td>2015</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Aerospace Engineering and Economics **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>201</th>\n",
       "      <td>2015</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Business and Computer Engineering **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>203</th>\n",
       "      <td>2015</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Chemical And Biomolecular Engineering and Econ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>209</th>\n",
       "      <td>2015</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Environmental Engineering and Economics **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>211</th>\n",
       "      <td>2015</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Information Engineering And Media and Economic...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>214</th>\n",
       "      <td>2015</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Mechanical Engineering and Economics **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>241</th>\n",
       "      <td>2015</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Computational Biology) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>242</th>\n",
       "      <td>2015</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Computer Engineering) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>266</th>\n",
       "      <td>2015</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Applied Science **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>270</th>\n",
       "      <td>2015</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Science (Computational Biology) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>288</th>\n",
       "      <td>2015</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Science (Architecture and Sustaina...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>289</th>\n",
       "      <td>2015</td>\n",
       "      <td>SIT</td>\n",
       "      <td>DigiPen Institute of Technology</td>\n",
       "      <td>Bachelor of Arts in Game Design **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>291</th>\n",
       "      <td>2015</td>\n",
       "      <td>SIT</td>\n",
       "      <td>DigiPen Institute of Technology</td>\n",
       "      <td>Bachelor of Science in Computer Science &amp; Game...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>308</th>\n",
       "      <td>2015</td>\n",
       "      <td>SIT</td>\n",
       "      <td>Trinity College Dublin</td>\n",
       "      <td>Bachelor in Science (Radiation Therapy) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>322</th>\n",
       "      <td>2016</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Aerospace Engineering and Economics **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>324</th>\n",
       "      <td>2016</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Business and Computer Engineering **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>334</th>\n",
       "      <td>2016</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Materials Engineering and Economics **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>336</th>\n",
       "      <td>2016</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Mechanical Engineering and Economics **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>345</th>\n",
       "      <td>2016</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Humanities, Arts &amp; Social Sciences</td>\n",
       "      <td>Public Policy And Global Affairs **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>371</th>\n",
       "      <td>2016</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Applied Science **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>375</th>\n",
       "      <td>2016</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Science (Computational Biology) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>381</th>\n",
       "      <td>2016</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Communications And Medi...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>383</th>\n",
       "      <td>2016</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Electronic Commerce) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>406</th>\n",
       "      <td>2016</td>\n",
       "      <td>SIT</td>\n",
       "      <td>DigiPen Institute of Technology</td>\n",
       "      <td>Bachelor of Arts in Game Design **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>425</th>\n",
       "      <td>2016</td>\n",
       "      <td>SIT</td>\n",
       "      <td>Trinity College Dublin / Singapore Institute o...</td>\n",
       "      <td>Bachelor in Science (Radiation Therapy) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>437</th>\n",
       "      <td>2016</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Science (Architecture and Sustaina...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>443</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Double Degree in Bachelor of Engineering (Hons...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>446</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Double Degree in Bachelor of Engineering  (Hon...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>448</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Double Degree in Bachelor of Engineering (Hons...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>451</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Double Degree in Bachelor of Business (Hons) a...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>455</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Double Degree in Bachelor of Engineering (Hons...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>458</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Double Degree in Bachelor of Engineering (Hons...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>460</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Double Degree in Bachelor of Engineering (Hons...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>468</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Engineering</td>\n",
       "      <td>Bachelor of Arts (Hons) in Philosophy **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>495</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Applied Science **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>499</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>Faculty of Science</td>\n",
       "      <td>Bachelor of Science (Computational Biology) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>505</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Communications And Medi...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>506</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Computational Biology) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>508</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Electronic Commerce) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>509</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>School of Computing</td>\n",
       "      <td>Bachelor of Computing (Information Security) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>552</th>\n",
       "      <td>2017</td>\n",
       "      <td>SIT</td>\n",
       "      <td>Singapore Institute of Technology -Trinity Col...</td>\n",
       "      <td>Bachelor in Science (Diagnostic Radiography) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>555</th>\n",
       "      <td>2017</td>\n",
       "      <td>SIT</td>\n",
       "      <td>Singapore Institute of Technology -Trinity Col...</td>\n",
       "      <td>Bachelor in Science (Radiation Therapy) **</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>569</th>\n",
       "      <td>2017</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Science (Architecture and Sustaina...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     year university                                             school  \\\n",
       "24   2013        NTU                                College of Sciences   \n",
       "28   2013        NTU                                College of Sciences   \n",
       "42   2013        NUS                                School of Computing   \n",
       "43   2013        NUS                                School of Computing   \n",
       "48   2013        NUS                     School of Design & Environment   \n",
       "66   2013        NUS                Yong Siew Toh Conservatory of Music   \n",
       "67   2013        NUS                                 Faculty of Science   \n",
       "71   2013        NUS                                 Faculty of Science   \n",
       "109  2014        NTU                                College of Sciences   \n",
       "127  2014        NUS                                School of Computing   \n",
       "128  2014        NUS                                School of Computing   \n",
       "151  2014        NUS                Yong Siew Toh Conservatory of Music   \n",
       "152  2014        NUS                                 Faculty of Science   \n",
       "156  2014        NUS                                 Faculty of Science   \n",
       "199  2015        NTU                             College of Engineering   \n",
       "201  2015        NTU                             College of Engineering   \n",
       "203  2015        NTU                             College of Engineering   \n",
       "209  2015        NTU                             College of Engineering   \n",
       "211  2015        NTU                             College of Engineering   \n",
       "214  2015        NTU                             College of Engineering   \n",
       "241  2015        NUS                                School of Computing   \n",
       "242  2015        NUS                                School of Computing   \n",
       "266  2015        NUS                                 Faculty of Science   \n",
       "270  2015        NUS                                 Faculty of Science   \n",
       "288  2015       SUTD                                                NaN   \n",
       "289  2015        SIT                    DigiPen Institute of Technology   \n",
       "291  2015        SIT                    DigiPen Institute of Technology   \n",
       "308  2015        SIT                             Trinity College Dublin   \n",
       "322  2016        NTU                             College of Engineering   \n",
       "324  2016        NTU                             College of Engineering   \n",
       "334  2016        NTU                             College of Engineering   \n",
       "336  2016        NTU                             College of Engineering   \n",
       "345  2016        NTU      College of Humanities, Arts & Social Sciences   \n",
       "371  2016        NUS                                 Faculty of Science   \n",
       "375  2016        NUS                                 Faculty of Science   \n",
       "381  2016        NUS                                School of Computing   \n",
       "383  2016        NUS                                School of Computing   \n",
       "406  2016        SIT                    DigiPen Institute of Technology   \n",
       "425  2016        SIT  Trinity College Dublin / Singapore Institute o...   \n",
       "437  2016       SUTD                                                NaN   \n",
       "443  2017        NTU                             College of Engineering   \n",
       "446  2017        NTU                             College of Engineering   \n",
       "448  2017        NTU                             College of Engineering   \n",
       "451  2017        NTU                             College of Engineering   \n",
       "455  2017        NTU                             College of Engineering   \n",
       "458  2017        NTU                             College of Engineering   \n",
       "460  2017        NTU                             College of Engineering   \n",
       "468  2017        NTU                             College of Engineering   \n",
       "495  2017        NUS                                 Faculty of Science   \n",
       "499  2017        NUS                                 Faculty of Science   \n",
       "505  2017        NUS                                School of Computing   \n",
       "506  2017        NUS                                School of Computing   \n",
       "508  2017        NUS                                School of Computing   \n",
       "509  2017        NUS                                School of Computing   \n",
       "552  2017        SIT  Singapore Institute of Technology -Trinity Col...   \n",
       "555  2017        SIT  Singapore Institute of Technology -Trinity Col...   \n",
       "569  2017       SUTD                                                NaN   \n",
       "\n",
       "                                                degree  \\\n",
       "24                              Biomedical Sciences **   \n",
       "28                          Mathematics & Economics **   \n",
       "42    Bachelor of Computing (Computational Biology) **   \n",
       "43     Bachelor of Computing (Computer Engineering) **   \n",
       "48               Bachelor of Arts (Architecture) ** ##   \n",
       "66                                Bachelor of Music **   \n",
       "67                      Bachelor of Applied Science **   \n",
       "71      Bachelor of Science (Computational Biology) **   \n",
       "109                              Biomedical Science **   \n",
       "127   Bachelor of Computing (Computational Biology) **   \n",
       "128    Bachelor of Computing (Computer Engineering) **   \n",
       "151                               Bachelor of Music **   \n",
       "152                     Bachelor of Applied Science **   \n",
       "156     Bachelor of Science (Computational Biology) **   \n",
       "199             Aerospace Engineering and Economics **   \n",
       "201               Business and Computer Engineering **   \n",
       "203  Chemical And Biomolecular Engineering and Econ...   \n",
       "209         Environmental Engineering and Economics **   \n",
       "211  Information Engineering And Media and Economic...   \n",
       "214            Mechanical Engineering and Economics **   \n",
       "241   Bachelor of Computing (Computational Biology) **   \n",
       "242    Bachelor of Computing (Computer Engineering) **   \n",
       "266                     Bachelor of Applied Science **   \n",
       "270     Bachelor of Science (Computational Biology) **   \n",
       "288  Bachelor of Science (Architecture and Sustaina...   \n",
       "289                 Bachelor of Arts in Game Design **   \n",
       "291  Bachelor of Science in Computer Science & Game...   \n",
       "308         Bachelor in Science (Radiation Therapy) **   \n",
       "322             Aerospace Engineering and Economics **   \n",
       "324               Business and Computer Engineering **   \n",
       "334             Materials Engineering and Economics **   \n",
       "336            Mechanical Engineering and Economics **   \n",
       "345                Public Policy And Global Affairs **   \n",
       "371                     Bachelor of Applied Science **   \n",
       "375     Bachelor of Science (Computational Biology) **   \n",
       "381  Bachelor of Computing (Communications And Medi...   \n",
       "383     Bachelor of Computing (Electronic Commerce) **   \n",
       "406                 Bachelor of Arts in Game Design **   \n",
       "425         Bachelor in Science (Radiation Therapy) **   \n",
       "437  Bachelor of Science (Architecture and Sustaina...   \n",
       "443  Double Degree in Bachelor of Engineering (Hons...   \n",
       "446  Double Degree in Bachelor of Engineering  (Hon...   \n",
       "448  Double Degree in Bachelor of Engineering (Hons...   \n",
       "451  Double Degree in Bachelor of Business (Hons) a...   \n",
       "455  Double Degree in Bachelor of Engineering (Hons...   \n",
       "458  Double Degree in Bachelor of Engineering (Hons...   \n",
       "460  Double Degree in Bachelor of Engineering (Hons...   \n",
       "468           Bachelor of Arts (Hons) in Philosophy **   \n",
       "495                     Bachelor of Applied Science **   \n",
       "499     Bachelor of Science (Computational Biology) **   \n",
       "505  Bachelor of Computing (Communications And Medi...   \n",
       "506   Bachelor of Computing (Computational Biology) **   \n",
       "508     Bachelor of Computing (Electronic Commerce) **   \n",
       "509    Bachelor of Computing (Information Security) **   \n",
       "552    Bachelor in Science (Diagnostic Radiography) **   \n",
       "555         Bachelor in Science (Radiation Therapy) **   \n",
       "569  Bachelor of Science (Architecture and Sustaina...   \n",
       "\n",
       "    employment_rate_overall employment_rate_ft_perm basic_monthly_mean  \\\n",
       "24                      NaN                     NaN                NaN   \n",
       "28                      NaN                     NaN                NaN   \n",
       "42                      NaN                     NaN                NaN   \n",
       "43                      NaN                     NaN                NaN   \n",
       "48                      NaN                     NaN                NaN   \n",
       "66                      NaN                     NaN                NaN   \n",
       "67                      NaN                     NaN                NaN   \n",
       "71                      NaN                     NaN                NaN   \n",
       "109                     NaN                     NaN                NaN   \n",
       "127                     NaN                     NaN                NaN   \n",
       "128                     NaN                     NaN                NaN   \n",
       "151                     NaN                     NaN                NaN   \n",
       "152                     NaN                     NaN                NaN   \n",
       "156                     NaN                     NaN                NaN   \n",
       "199                     NaN                     NaN                NaN   \n",
       "201                     NaN                     NaN                NaN   \n",
       "203                     NaN                     NaN                NaN   \n",
       "209                     NaN                     NaN                NaN   \n",
       "211                     NaN                     NaN                NaN   \n",
       "214                     NaN                     NaN                NaN   \n",
       "241                     NaN                     NaN                NaN   \n",
       "242                     NaN                     NaN                NaN   \n",
       "266                     NaN                     NaN                NaN   \n",
       "270                     NaN                     NaN                NaN   \n",
       "288                     NaN                     NaN                NaN   \n",
       "289                     NaN                     NaN                NaN   \n",
       "291                     NaN                     NaN                NaN   \n",
       "308                     NaN                     NaN                NaN   \n",
       "322                     NaN                     NaN                NaN   \n",
       "324                     NaN                     NaN                NaN   \n",
       "334                     NaN                     NaN                NaN   \n",
       "336                     NaN                     NaN                NaN   \n",
       "345                     NaN                     NaN                NaN   \n",
       "371                     NaN                     NaN                NaN   \n",
       "375                     NaN                     NaN                NaN   \n",
       "381                     NaN                     NaN                NaN   \n",
       "383                     NaN                     NaN                NaN   \n",
       "406                     NaN                     NaN                NaN   \n",
       "425                     NaN                     NaN                NaN   \n",
       "437                     NaN                     NaN                NaN   \n",
       "443                     NaN                     NaN                NaN   \n",
       "446                     NaN                     NaN                NaN   \n",
       "448                     NaN                     NaN                NaN   \n",
       "451                     NaN                     NaN                NaN   \n",
       "455                     NaN                     NaN                NaN   \n",
       "458                     NaN                     NaN                NaN   \n",
       "460                     NaN                     NaN                NaN   \n",
       "468                     NaN                     NaN                NaN   \n",
       "495                     NaN                     NaN                NaN   \n",
       "499                     NaN                     NaN                NaN   \n",
       "505                     NaN                     NaN                NaN   \n",
       "506                     NaN                     NaN                NaN   \n",
       "508                     NaN                     NaN                NaN   \n",
       "509                     NaN                     NaN                NaN   \n",
       "552                     NaN                     NaN                NaN   \n",
       "555                     NaN                     NaN                NaN   \n",
       "569                     NaN                     NaN                NaN   \n",
       "\n",
       "    basic_monthly_median gross_monthly_mean gross_monthly_median  \\\n",
       "24                   NaN                NaN                  NaN   \n",
       "28                   NaN                NaN                  NaN   \n",
       "42                   NaN                NaN                  NaN   \n",
       "43                   NaN                NaN                  NaN   \n",
       "48                   NaN                NaN                  NaN   \n",
       "66                   NaN                NaN                  NaN   \n",
       "67                   NaN                NaN                  NaN   \n",
       "71                   NaN                NaN                  NaN   \n",
       "109                  NaN                NaN                  NaN   \n",
       "127                  NaN                NaN                  NaN   \n",
       "128                  NaN                NaN                  NaN   \n",
       "151                  NaN                NaN                  NaN   \n",
       "152                  NaN                NaN                  NaN   \n",
       "156                  NaN                NaN                  NaN   \n",
       "199                  NaN                NaN                  NaN   \n",
       "201                  NaN                NaN                  NaN   \n",
       "203                  NaN                NaN                  NaN   \n",
       "209                  NaN                NaN                  NaN   \n",
       "211                  NaN                NaN                  NaN   \n",
       "214                  NaN                NaN                  NaN   \n",
       "241                  NaN                NaN                  NaN   \n",
       "242                  NaN                NaN                  NaN   \n",
       "266                  NaN                NaN                  NaN   \n",
       "270                  NaN                NaN                  NaN   \n",
       "288                  NaN                NaN                  NaN   \n",
       "289                  NaN                NaN                  NaN   \n",
       "291                  NaN                NaN                  NaN   \n",
       "308                  NaN                NaN                  NaN   \n",
       "322                  NaN                NaN                  NaN   \n",
       "324                  NaN                NaN                  NaN   \n",
       "334                  NaN                NaN                  NaN   \n",
       "336                  NaN                NaN                  NaN   \n",
       "345                  NaN                NaN                  NaN   \n",
       "371                  NaN                NaN                  NaN   \n",
       "375                  NaN                NaN                  NaN   \n",
       "381                  NaN                NaN                  NaN   \n",
       "383                  NaN                NaN                  NaN   \n",
       "406                  NaN                NaN                  NaN   \n",
       "425                  NaN                NaN                  NaN   \n",
       "437                  NaN                NaN                  NaN   \n",
       "443                  NaN                NaN                  NaN   \n",
       "446                  NaN                NaN                  NaN   \n",
       "448                  NaN                NaN                  NaN   \n",
       "451                  NaN                NaN                  NaN   \n",
       "455                  NaN                NaN                  NaN   \n",
       "458                  NaN                NaN                  NaN   \n",
       "460                  NaN                NaN                  NaN   \n",
       "468                  NaN                NaN                  NaN   \n",
       "495                  NaN                NaN                  NaN   \n",
       "499                  NaN                NaN                  NaN   \n",
       "505                  NaN                NaN                  NaN   \n",
       "506                  NaN                NaN                  NaN   \n",
       "508                  NaN                NaN                  NaN   \n",
       "509                  NaN                NaN                  NaN   \n",
       "552                  NaN                NaN                  NaN   \n",
       "555                  NaN                NaN                  NaN   \n",
       "569                  NaN                NaN                  NaN   \n",
       "\n",
       "    gross_mthly_25_percentile gross_mthly_75_percentile  \n",
       "24                        NaN                       NaN  \n",
       "28                        NaN                       NaN  \n",
       "42                        NaN                       NaN  \n",
       "43                        NaN                       NaN  \n",
       "48                        NaN                       NaN  \n",
       "66                        NaN                       NaN  \n",
       "67                        NaN                       NaN  \n",
       "71                        NaN                       NaN  \n",
       "109                       NaN                       NaN  \n",
       "127                       NaN                       NaN  \n",
       "128                       NaN                       NaN  \n",
       "151                       NaN                       NaN  \n",
       "152                       NaN                       NaN  \n",
       "156                       NaN                       NaN  \n",
       "199                       NaN                       NaN  \n",
       "201                       NaN                       NaN  \n",
       "203                       NaN                       NaN  \n",
       "209                       NaN                       NaN  \n",
       "211                       NaN                       NaN  \n",
       "214                       NaN                       NaN  \n",
       "241                       NaN                       NaN  \n",
       "242                       NaN                       NaN  \n",
       "266                       NaN                       NaN  \n",
       "270                       NaN                       NaN  \n",
       "288                       NaN                       NaN  \n",
       "289                       NaN                       NaN  \n",
       "291                       NaN                       NaN  \n",
       "308                       NaN                       NaN  \n",
       "322                       NaN                       NaN  \n",
       "324                       NaN                       NaN  \n",
       "334                       NaN                       NaN  \n",
       "336                       NaN                       NaN  \n",
       "345                       NaN                       NaN  \n",
       "371                       NaN                       NaN  \n",
       "375                       NaN                       NaN  \n",
       "381                       NaN                       NaN  \n",
       "383                       NaN                       NaN  \n",
       "406                       NaN                       NaN  \n",
       "425                       NaN                       NaN  \n",
       "437                       NaN                       NaN  \n",
       "443                       NaN                       NaN  \n",
       "446                       NaN                       NaN  \n",
       "448                       NaN                       NaN  \n",
       "451                       NaN                       NaN  \n",
       "455                       NaN                       NaN  \n",
       "458                       NaN                       NaN  \n",
       "460                       NaN                       NaN  \n",
       "468                       NaN                       NaN  \n",
       "495                       NaN                       NaN  \n",
       "499                       NaN                       NaN  \n",
       "505                       NaN                       NaN  \n",
       "506                       NaN                       NaN  \n",
       "508                       NaN                       NaN  \n",
       "509                       NaN                       NaN  \n",
       "552                       NaN                       NaN  \n",
       "555                       NaN                       NaN  \n",
       "569                       NaN                       NaN  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# It can be seen that if employment_rate_overall is NaN, the remaining features would be blank as well.\n",
    "# This makes sense because every column after employment_rate_overall is dependent on whether if there is any employment\n",
    "\n",
    "df[df['employment_rate_overall'].isnull() == True]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 513 entries, 0 to 568\n",
      "Data columns (total 12 columns):\n",
      "year                         513 non-null int64\n",
      "university                   513 non-null object\n",
      "school                       504 non-null object\n",
      "degree                       513 non-null object\n",
      "employment_rate_overall      513 non-null object\n",
      "employment_rate_ft_perm      513 non-null object\n",
      "basic_monthly_mean           513 non-null object\n",
      "basic_monthly_median         513 non-null object\n",
      "gross_monthly_mean           513 non-null object\n",
      "gross_monthly_median         513 non-null object\n",
      "gross_mthly_25_percentile    513 non-null object\n",
      "gross_mthly_75_percentile    513 non-null object\n",
      "dtypes: int64(1), object(11)\n",
      "memory usage: 52.1+ KB\n"
     ]
    }
   ],
   "source": [
    "# Remove rows where employment_rate_overall is NaN since those rows are irrelevant\n",
    "df = df[df['employment_rate_overall'].isnull() == False]\n",
    "\n",
    "# Check to see if NaN rows has been successfully removed\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "scrolled": true
   },
   "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>university</th>\n",
       "      <th>school</th>\n",
       "      <th>degree</th>\n",
       "      <th>employment_rate_overall</th>\n",
       "      <th>employment_rate_ft_perm</th>\n",
       "      <th>basic_monthly_mean</th>\n",
       "      <th>basic_monthly_median</th>\n",
       "      <th>gross_monthly_mean</th>\n",
       "      <th>gross_monthly_median</th>\n",
       "      <th>gross_mthly_25_percentile</th>\n",
       "      <th>gross_mthly_75_percentile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>285</th>\n",
       "      <td>2015</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Engineering (Engineering Product D...</td>\n",
       "      <td>92.5</td>\n",
       "      <td>88.1</td>\n",
       "      <td>3595</td>\n",
       "      <td>3618</td>\n",
       "      <td>3565</td>\n",
       "      <td>3600</td>\n",
       "      <td>3400</td>\n",
       "      <td>3800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>286</th>\n",
       "      <td>2015</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Engineering (Engineering Systems a...</td>\n",
       "      <td>80.3</td>\n",
       "      <td>71.8</td>\n",
       "      <td>3699</td>\n",
       "      <td>3450</td>\n",
       "      <td>3763</td>\n",
       "      <td>3500</td>\n",
       "      <td>3190</td>\n",
       "      <td>4100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>287</th>\n",
       "      <td>2015</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Engineering (Information Systems T...</td>\n",
       "      <td>86</td>\n",
       "      <td>83.7</td>\n",
       "      <td>3872</td>\n",
       "      <td>3700</td>\n",
       "      <td>3952</td>\n",
       "      <td>3950</td>\n",
       "      <td>3338</td>\n",
       "      <td>4450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>434</th>\n",
       "      <td>2016</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Engineering (Engineering Product D...</td>\n",
       "      <td>87</td>\n",
       "      <td>79.7</td>\n",
       "      <td>3620</td>\n",
       "      <td>3500</td>\n",
       "      <td>3690</td>\n",
       "      <td>3600</td>\n",
       "      <td>3300</td>\n",
       "      <td>3750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>435</th>\n",
       "      <td>2016</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Engineering (Information Systems T...</td>\n",
       "      <td>91.9</td>\n",
       "      <td>91.9</td>\n",
       "      <td>4197</td>\n",
       "      <td>4000</td>\n",
       "      <td>4232</td>\n",
       "      <td>4000</td>\n",
       "      <td>3500</td>\n",
       "      <td>4950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>436</th>\n",
       "      <td>2016</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Engineering (Engineering Systems a...</td>\n",
       "      <td>100</td>\n",
       "      <td>93.5</td>\n",
       "      <td>3695</td>\n",
       "      <td>3600</td>\n",
       "      <td>3830</td>\n",
       "      <td>3650</td>\n",
       "      <td>3550</td>\n",
       "      <td>4200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>566</th>\n",
       "      <td>2017</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Engineering (Engineering Product D...</td>\n",
       "      <td>89.9</td>\n",
       "      <td>87.3</td>\n",
       "      <td>3594</td>\n",
       "      <td>3600</td>\n",
       "      <td>3638</td>\n",
       "      <td>3600</td>\n",
       "      <td>3450</td>\n",
       "      <td>3800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>567</th>\n",
       "      <td>2017</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Engineering (Information Systems T...</td>\n",
       "      <td>100</td>\n",
       "      <td>100</td>\n",
       "      <td>4146</td>\n",
       "      <td>4000</td>\n",
       "      <td>4175</td>\n",
       "      <td>4023</td>\n",
       "      <td>3600</td>\n",
       "      <td>4900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>568</th>\n",
       "      <td>2017</td>\n",
       "      <td>SUTD</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bachelor of Engineering (Engineering Systems a...</td>\n",
       "      <td>91.7</td>\n",
       "      <td>85.4</td>\n",
       "      <td>3856</td>\n",
       "      <td>3800</td>\n",
       "      <td>3956</td>\n",
       "      <td>3850</td>\n",
       "      <td>3600</td>\n",
       "      <td>4210</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     year university school  \\\n",
       "285  2015       SUTD    NaN   \n",
       "286  2015       SUTD    NaN   \n",
       "287  2015       SUTD    NaN   \n",
       "434  2016       SUTD    NaN   \n",
       "435  2016       SUTD    NaN   \n",
       "436  2016       SUTD    NaN   \n",
       "566  2017       SUTD    NaN   \n",
       "567  2017       SUTD    NaN   \n",
       "568  2017       SUTD    NaN   \n",
       "\n",
       "                                                degree  \\\n",
       "285  Bachelor of Engineering (Engineering Product D...   \n",
       "286  Bachelor of Engineering (Engineering Systems a...   \n",
       "287  Bachelor of Engineering (Information Systems T...   \n",
       "434  Bachelor of Engineering (Engineering Product D...   \n",
       "435  Bachelor of Engineering (Information Systems T...   \n",
       "436  Bachelor of Engineering (Engineering Systems a...   \n",
       "566  Bachelor of Engineering (Engineering Product D...   \n",
       "567  Bachelor of Engineering (Information Systems T...   \n",
       "568  Bachelor of Engineering (Engineering Systems a...   \n",
       "\n",
       "    employment_rate_overall employment_rate_ft_perm basic_monthly_mean  \\\n",
       "285                    92.5                    88.1               3595   \n",
       "286                    80.3                    71.8               3699   \n",
       "287                      86                    83.7               3872   \n",
       "434                      87                    79.7               3620   \n",
       "435                    91.9                    91.9               4197   \n",
       "436                     100                    93.5               3695   \n",
       "566                    89.9                    87.3               3594   \n",
       "567                     100                     100               4146   \n",
       "568                    91.7                    85.4               3856   \n",
       "\n",
       "    basic_monthly_median gross_monthly_mean gross_monthly_median  \\\n",
       "285                 3618               3565                 3600   \n",
       "286                 3450               3763                 3500   \n",
       "287                 3700               3952                 3950   \n",
       "434                 3500               3690                 3600   \n",
       "435                 4000               4232                 4000   \n",
       "436                 3600               3830                 3650   \n",
       "566                 3600               3638                 3600   \n",
       "567                 4000               4175                 4023   \n",
       "568                 3800               3956                 3850   \n",
       "\n",
       "    gross_mthly_25_percentile gross_mthly_75_percentile  \n",
       "285                      3400                      3800  \n",
       "286                      3190                      4100  \n",
       "287                      3338                      4450  \n",
       "434                      3300                      3750  \n",
       "435                      3500                      4950  \n",
       "436                      3550                      4200  \n",
       "566                      3450                      3800  \n",
       "567                      3600                      4900  \n",
       "568                      3600                      4210  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We do not have to remove NaN from school as they are all belonged to SUTD. NaN in this case simply suggest a lack of a suitable school name.\n",
    "# For this exercise, I did not include SUTD as the university does not have Accountancy and Business\n",
    "# However, for readers who wish to take my code and analyse the bachelor of engineering, you will have to rename NaN as school of engineering perhaps\n",
    "\n",
    "df[df['university'] == 'SUTD']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 513 entries, 0 to 568\n",
      "Data columns (total 12 columns):\n",
      "year                         513 non-null int64\n",
      "university                   513 non-null object\n",
      "school                       513 non-null object\n",
      "degree                       513 non-null object\n",
      "employment_rate_overall      513 non-null float64\n",
      "employment_rate_ft_perm      513 non-null float64\n",
      "basic_monthly_mean           513 non-null int64\n",
      "basic_monthly_median         513 non-null int64\n",
      "gross_monthly_mean           513 non-null int64\n",
      "gross_monthly_median         513 non-null int64\n",
      "gross_mthly_25_percentile    513 non-null int64\n",
      "gross_mthly_75_percentile    513 non-null int64\n",
      "dtypes: float64(2), int64(7), object(3)\n",
      "memory usage: 52.1+ KB\n"
     ]
    }
   ],
   "source": [
    "# Convert variable types \n",
    "df[['employment_rate_overall', 'employment_rate_ft_perm']] = df[['employment_rate_overall', 'employment_rate_ft_perm']].astype(float)\n",
    "\n",
    "# Might consider using a loop to convert to the code shorter\n",
    "df[['basic_monthly_mean', 'basic_monthly_median', \n",
    "    'gross_monthly_mean', 'gross_monthly_median', \n",
    "    'gross_mthly_25_percentile', \n",
    "    'gross_mthly_75_percentile']] = df[['basic_monthly_mean', 'basic_monthly_median', \n",
    "                                        'gross_monthly_mean', 'gross_monthly_median',\n",
    "                                        'gross_mthly_25_percentile', 'gross_mthly_75_percentile']].astype('int64')\n",
    "\n",
    "# Check to ensure the data types have been changed successfully \n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NTU ['Accountancy and Business' 'Accountancy (3-yr direct Honours Programme)'\n",
      " 'Business (3-yr direct Honours Programme)' 'Business and Computing'\n",
      " 'Aerospace Engineering' 'Bioengineering'\n",
      " 'Chemical and Biomolecular Engineering' 'Computer Engineering'\n",
      " 'Civil Engineering' 'Computer Science'\n",
      " 'Electrical and Electronic Engineering' 'Environmental Engineering'\n",
      " 'Information Engineering and Media' 'Materials Engineering'\n",
      " 'Mechanical Engineering' 'Maritime Studies' 'Art, Design & Media'\n",
      " 'Chinese' 'Communication Studies' 'Economics' 'English'\n",
      " 'Linguistics and Multilingual Studies' 'Psychology' 'Sociology'\n",
      " 'Biomedical Sciences (Traditional Chinese Medicine) ^'\n",
      " 'Biological Sciences' 'Chemistry & Biological Chemistry'\n",
      " 'Mathematical Science' 'Physics / Applied Physics'\n",
      " 'Sports Science and Management' 'Science (with Education)'\n",
      " 'Arts (with Education)'\n",
      " 'Biomedical Science (Traditional Chinese Medicine) ^'\n",
      " 'Mathematics & Economics' 'Chemical And Biomolecular Engineering'\n",
      " 'Electrical And Electronic Engineering'\n",
      " 'Information Engineering And Media'\n",
      " 'Linguistics And Multilingual Studies' 'Accountancy' 'Business' 'History'\n",
      " 'Biomedical Sciences & Chinese Medicine ^' 'Mathematical Sciences'\n",
      " 'Bachelor of Accountancy (Hons)'\n",
      " 'Double Degree in Bachelor of Accountancy (Hons) and Bachelor of Business (Hons)'\n",
      " 'Bachelor of Business (Hons)'\n",
      " 'Double Degree in Bachelor of Business (Hons) and Bachelor of Engineering (Hons) (Computer Science)'\n",
      " 'Bachelor of Engineering (Hons) (Aerospace Engineering)'\n",
      " 'Bachelor of Engineering (Hons) (Bioengineering)'\n",
      " 'Bachelor of Engineering  (Hons) (Chemical And Biomolecular Engineering)'\n",
      " 'Bachelor of Engineering (Hons) (Civil Engineering)'\n",
      " 'Bachelor of Engineering (Hons) (Computer Engineering)'\n",
      " 'Bachelor of Engineering (Hons) (Computer Science)'\n",
      " 'Bachelor of Engineering (Hons) (Electrical & Electronic Engineering)'\n",
      " 'Bachelor of Engineering (Hons) (Environmental Engineering)'\n",
      " 'Bachelor of Engineering (Hons) (Information Engineering & Media)'\n",
      " 'Bachelor of Science (Hons) (Maritime Studies)'\n",
      " 'Bachelor of Engineering (Hons) (Materials Engineering)'\n",
      " 'Bachelor of Engineering (Hons) (Mechanical Engineering)'\n",
      " 'Bachelor of Fine Arts (Hons)' 'Bachelor of Arts (Hons) in Chinese'\n",
      " 'Bachelor of Communication Studies (Hons)'\n",
      " 'Bachelor of Arts (Hons) in Economics'\n",
      " 'Bachelor of Arts (Hons) in English' 'Bachelor of Arts (Hons) in History'\n",
      " 'Bachelor of Arts (Hons) in Linguistics And Multilingual Studies'\n",
      " 'Bachelor of Arts (Hons) in Psychology'\n",
      " 'Bachelor of Arts (Hons) in Public Policy And Global Affairs'\n",
      " 'Bachelor of Arts (Hons) in Sociology'\n",
      " 'Double Degree in Bachelor of Science (Hons) in Biomedical Sciences and Bachelor of Medicine (Chinese Medicine) ^'\n",
      " 'Bachelor of Science (Hons) in Biological Sciences'\n",
      " 'Bachelor of Science (Hons) in Chemistry & Biological Chemistry'\n",
      " 'Bachelor of Science (Hons) in Mathematical Sciences'\n",
      " 'Bachelor of Science (Hons) in Mathematics & Economics'\n",
      " 'Bachelor of Science (Hons) in Physics/Applied Physics'\n",
      " 'Bachelor of Science (Hons) (Sport Science & Management)'\n",
      " 'Bachelor of Arts (Hons) (Education)'\n",
      " 'Bachelor of Science (Hons) (Education)'] \n",
      "\n",
      "NUS ['Bachelor of Arts' 'Bachelor of Arts (Hons)'\n",
      " 'Bachelor of Social Sciences' 'Bachelor of Business Administration'\n",
      " 'Bachelor of Business Administration (Hons)'\n",
      " 'Bachelor of Business Administration (Accountancy)'\n",
      " 'Bachelor of Business Administration (Accountancy) (Hons)'\n",
      " 'Bachelor of Computing (Communications and Media)'\n",
      " 'Bachelor of Computing (Computer Science)'\n",
      " 'Bachelor of Computing (Electronic Commerce)'\n",
      " 'Bachelor of Computing (Information Systems)'\n",
      " 'Bachelor of Dental Surgery' 'Bachelor of Arts (Industrial Design)'\n",
      " 'Bachelor of Science (Project and Facilities Management)'\n",
      " 'Bachelor of Science (Real Estate)'\n",
      " 'Bachelor of Engineering (Bioengineering)'\n",
      " 'Bachelor of Engineering (Chemical Engineering)'\n",
      " 'Bachelor of Engineering (Civil Engineering)'\n",
      " 'Bachelor of Engineering (Computer Engineering)'\n",
      " 'Bachelor of Engineering (Electrical Engineering)'\n",
      " 'Bachelor of Engineering (Engineering Science)'\n",
      " 'Bachelor of Engineering (Environmental Engineering)'\n",
      " 'Bachelor of Engineering (Industrial and Systems Engineering)'\n",
      " 'Bachelor of Engineering (Materials Science and Engineering)'\n",
      " 'Bachelor of Engineering (Mechanical Engineering)'\n",
      " 'Bachelor of Laws (LLB) (Hons) ##'\n",
      " 'Bachelor of Medicine and Bachelor of Surgery (MBBS) ##'\n",
      " 'Bachelor of Science (Nursing)' 'Bachelor of Science (Nursing) (Hons)'\n",
      " 'Bachelor of Applied Science (Hons)' 'Bachelor of Science'\n",
      " 'Bachelor of Science (Hons)' 'Bachelor of Science (Pharmacy) (Hons) ##'\n",
      " 'Bachelor of Arts (Architecture) ##'\n",
      " 'Bachelor of Engineering (Biomedical Engineering) ^^'\n",
      " 'Bachelor of Medicine and Bachelor of Surgery ##'\n",
      " 'Bachelor of Engineering (Biomedical Engineering)'\n",
      " 'Bachelor of Laws (L.L.B) (Hons) ##' 'Bachelor of Music'\n",
      " 'Bachelor of Environmental Studies'\n",
      " 'Bachelor of Engineering (Materials Science And Engineering)'\n",
      " 'Bachelor of Science (Pharmacy)(Hons) ##'\n",
      " 'Bachelor of Medicine and Bachelor Of Surgery ##' 'Bachelor of Laws ##'\n",
      " 'Bachelor of Science (Pharmacy) ##'\n",
      " 'Bachelor of Science (Business Analytics)'\n",
      " 'Bachelor of Medicine And Bachelor Of Surgery ##'\n",
      " 'Bachelor of Arts with Honours' 'Bachelor of Science with Honours'] \n",
      "\n",
      "SMU ['Accountancy (4-years programme)'\n",
      " 'Accountancy (4-years programme) Cum Laude and above'\n",
      " 'Business Management (4-years programme)'\n",
      " 'Business Management (4-years programme) Cum Laude and above'\n",
      " 'Economics (4-years programme)'\n",
      " 'Economics (4-years programme) Cum Laude and above'\n",
      " 'Information Systems Management (4-years programme)'\n",
      " 'Information Systems Management (4-years programme) Cum Laude and above'\n",
      " 'Social Sciences (4-years programme)'\n",
      " 'Social Sciences (4-years programme) Cum Laude and above'\n",
      " 'Law (4-years programme) ++'\n",
      " 'Law (4-years programme) Cum Laude and above'\n",
      " 'Accountancy (4-year programme)'\n",
      " 'Accountancy (4-year programme) Cum Laude and above'\n",
      " 'Business Management (4-year programme)'\n",
      " 'Business Management (4-year programme) Cum Laude and above'\n",
      " 'Economics (4-year programme)'\n",
      " 'Economics (4-year programme) Cum Laude and above'\n",
      " 'Information Systems Management (4-year programme)'\n",
      " 'Information Systems Management (4-year programme) Cum Laude and above'\n",
      " 'Social Sciences (4-year programme)'\n",
      " 'Social Sciences (4-year programme) Cum Laude and above'\n",
      " 'Law (4-year programme) ++' 'Law (4-year programme) Cum Laude and above'\n",
      " 'Accountancy' 'Accountancy Cum Laude and above' 'Business Management'\n",
      " 'Business Management Cum Laude and above' 'Economics'\n",
      " 'Economics Cum Laude and above' 'Information Systems Management'\n",
      " 'Information Systems Management Cum Laude and above' 'Social Sciences'\n",
      " 'Social Sciences Cum Laude and above' 'Law ++' 'Law Cum Laude and above'] \n",
      "\n",
      "SIT ['Bachelor of Arts in Game Design'\n",
      " 'Bachelor of Fine Arts in Digital Arts & Animation'\n",
      " 'Bachelor of Science in Computer Science & Game Design'\n",
      " 'Bachelor of Science in Computer Science in Real-Time Interactive Simulation'\n",
      " 'Bachelor of Arts with Honours in Communication Design'\n",
      " 'Bachelor of Arts with Honours in Interior Design'\n",
      " 'Bachelor of Engineering with Honours in Chemical Engineering'\n",
      " 'Bachelor of Engineering with Honours in Marine Engineering'\n",
      " 'Bachelor of Engineering with Honours in Mechanical Design & Manufacturing Engineering'\n",
      " 'Bachelor of Engineering with Honours in Naval Architecture'\n",
      " 'Bachelor of Engineering with Honours in Offshore Engineering'\n",
      " 'Bachelor of Science with Honours in Food & Human Nutrition'\n",
      " 'Bachelor of Science in Chemical Engineering'\n",
      " 'Bachelor of Science in Electrical Engineering & Information Technology'\n",
      " 'Bachelor of Professional Studies in Culinary Arts Management'\n",
      " 'Bachelor in Science (Occupational Therapy)'\n",
      " 'Bachelor in Science (Physiotherapy)'\n",
      " 'Bachelor of Engineering with Honours in Aeronautical Engineering'\n",
      " 'Bachelor of Engineering with Honours in Aerospace Systems'\n",
      " 'Bachelor of Engineering with Honours in Mechanical Design Engineering'\n",
      " 'Bachelor of Engineering with Honours in Mechatronics'\n",
      " 'Bachelor of Science with Honours in Nursing Practice'\n",
      " 'Bachelor of Science (Major in Hospitality Management)'\n",
      " 'Bachelor of Science in Early Childhood Education'\n",
      " 'Bachelor of Engineering with Honours in Electrical Power Engineering'\n",
      " 'Bachelor in Science (Diagnostic Radiography)'\n",
      " 'Bachelor of Science with Honours in Computing Science'\n",
      " 'Bachelor of Arts with Honours in Criminology and Security'\n",
      " 'Bachelor of Arts in Game Design ^^^'\n",
      " 'Bachelor of Fine Arts in Digital Arts & Animation ^^^'\n",
      " 'Bachelor of Science in Computer Science & Game Design ^^^'\n",
      " 'Bachelor of Engineering with Honours in Marine Engineering ^^^'\n",
      " 'Bachelor of Engineering with Honours in Naval Architecture ^^^'\n",
      " 'Bachelor of Engineering with Honours in Offshore Engineering ^^^'\n",
      " 'Bachelor of Accountancy with Honours'\n",
      " 'Bachelor of Hospitality Business with Honours ^^^'] \n",
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "SUTD ['Bachelor of Engineering (Engineering Product Development)'\n",
      " 'Bachelor of Engineering (Engineering Systems and Design)'\n",
      " 'Bachelor of Engineering (Information Systems Technology and Design)'] \n",
      "\n"
     ]
    }
   ],
   "source": [
    "# A quick check of unique degree names to ensure no one degree is named differently. \n",
    "# Notice that for NTU, Accountancy (3-yr direct Honours Programme) is named differently from Bachelor of Accountancy (Hons)\n",
    "# This means that we will need to deal with the different naming convention later\n",
    "\n",
    "for uni in df['university'].unique():\n",
    "    print(str(uni), df[df['university'] == uni].degree.unique(), \"\\n\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'College of Business (Nanyang Business School)': 'NTU',\n",
       " 'College of Engineering': 'NTU',\n",
       " 'College of Humanities, Arts & Social Sciences': 'NTU',\n",
       " 'College of Sciences': 'NTU',\n",
       " 'National Institute of Education (NIE)': 'NTU',\n",
       " 'Faculty of Arts & Social Sciences': 'NUS',\n",
       " 'NUS Business School': 'NUS',\n",
       " 'School of Computing': 'NUS',\n",
       " 'Faculty of Dentistry': 'NUS',\n",
       " 'School of Design & Environment': 'NUS',\n",
       " 'Faculty of Engineering': 'NUS',\n",
       " 'Faculty of Law': 'NUS',\n",
       " 'YLL School of Medicine': 'NUS',\n",
       " 'Faculty of Science': 'NUS',\n",
       " 'School of Accountancy (4-years programme) *': 'SMU',\n",
       " 'School of Business (4-years programme) *': 'SMU',\n",
       " 'School of Economics (4-years programme) *': 'SMU',\n",
       " 'School of Information Systems (4-years programme) *': 'SMU',\n",
       " 'School of Social Sciences (4-years programme) *': 'SMU',\n",
       " 'School of Law (4-years programme) *': 'SMU',\n",
       " 'School of Accountancy (4-year programme) *': 'SMU',\n",
       " 'School of Business (4-year programme) *': 'SMU',\n",
       " 'School of Economics (4-year programme) *': 'SMU',\n",
       " 'School of Information Systems (4-year programme) *': 'SMU',\n",
       " 'School of Social Sciences (4-year programme) *': 'SMU',\n",
       " 'School of Law (4-year programme) *': 'SMU',\n",
       " 'DigiPen Institute of Technology': 'SIT',\n",
       " 'The Glasgow School of Art': 'SIT',\n",
       " 'Newcastle University': 'SIT',\n",
       " 'Technische Universität München': 'SIT',\n",
       " 'The Culinary Institute of America': 'SIT',\n",
       " 'Trinity College Dublin': 'SIT',\n",
       " 'University of Glasgow': 'SIT',\n",
       " 'University of Manchester': 'SIT',\n",
       " 'University of Nevada, Las Vegas': 'SIT',\n",
       " 'Wheelock College': 'SIT',\n",
       " 'Yong Siew Toh Conservatory of Music': 'NUS',\n",
       " 'Multidisciplinary Program': 'NUS',\n",
       " 'SUTD Engineering': 'SUTD',\n",
       " 'Singapore Institute of Technology -Trinity College Dublin': 'SIT',\n",
       " 'Sports Science and Management': 'NTU',\n",
       " 'Faculty Of Dentistry': 'NUS',\n",
       " 'Faculty Of Engineering': 'NUS',\n",
       " 'School of Design and Environment': 'NUS',\n",
       " 'Yong Loo Lin School (Medicine)': 'NUS',\n",
       " 'Multidisciplinary Programme': 'NUS',\n",
       " 'Trinity College Dublin / Singapore Institute of Technology-Trinity College Dublin': 'SIT',\n",
       " 'University of Liverpool': 'SIT',\n",
       " 'Sport Science and Management': 'NTU',\n",
       " 'YST Conservatory Of Music': 'NUS',\n",
       " 'Yale-NUS College': 'NUS',\n",
       " 'Multi-Disciplinary Programme': 'NUS',\n",
       " 'Singapore Institute of Technology -Trinity College Dublin / Trinity College Dublin': 'SIT',\n",
       " 'Singapore Institute of Technology': 'SIT'}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# We can use a dictionary, we can perform a check to see if a given business school \n",
    "# Notice that SMU is the only school that differentiates Business and Accountancy at a school level. This is ok as we are not combining Accountancy and Business\n",
    "\n",
    "dict(zip(df['school'], df['university']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## The reason why I choose not to compare between universities is due to a couple of issues.\n",
    "\n",
    "1) Insights are not meaningful. There is little takeaway in knowing how business schools as a whole performed in each university due to multiple variables (Explained below).\n",
    "\n",
    "2) Granularity. Each business school in general can be split into Business and Accountancy. Business degree contains multiple majors ranging from HR, Marketing, Business Analytics (NTU), Banking & Finance etc which the dataset does not contain such level of information.\n",
    "\n",
    "3) Single vs double degree. For this dataset, only NTU split between single and double degree. For SMU, double degree has been classified under as single degree using its first degree. While NUS and SIT only have single degree information. We can easily remove the double degree from NTU but not SMU as it has already been clubbed together. This makes the comparison inconsistent.\n",
    "\n",
    "4) Honours. Most of the business degrees have direct honours with the exception of NUS. Again, using only direct honours would result in an inconsistent comparison. On the other hand, taking the mean of the two might be risky because we do not know the exact sample size for each group (See proof below). \n",
    "\n",
    "5) SIT. SIT business degree is specific to hospitality. NTU for example, do not have such a major. Comparing at university level would mean that we are comparing the employment rates and salary of different majors which is an apple to orange comparison."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Proof as to why we cannot simply take the mean salary of hons and no hons to be the average of both"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "mean_employment_rate_unequal: 0.75\n",
      "Actual employment_rate: 0.7714285714285715\n"
     ]
    }
   ],
   "source": [
    "# Proof of how mean cannot be used if sample size is unknown\n",
    "\n",
    "# Assume equal sample size for both hons and non-hons\n",
    "df_proof_equal = pd.DataFrame({'Honors': ['Non-Hons', 'Hons'],\n",
    "                         'Sample_size': [100, 100],\n",
    "                         'Employment_rate': [0.7, 0.8],\n",
    "                         'Mean_salary': [3200, 3500]})\n",
    "\n",
    "df_proof_equal\n",
    "\n",
    "mean_employment_rate_equal = (0.7 + 0.8) / 2 # This is the same as (100*0.7 + 100 * 0.8)/(100+100)\n",
    "\n",
    "# However, it is possible that the sample size is not equal.\n",
    "df_proof_unequal = pd.DataFrame({'Honors': ['Non-Hons', 'Hons'],\n",
    "                         'Sample_size': [100, 250], # 2.5 times is just an arbitrary number. Hons is certainly more than non-hons though\n",
    "                         'Employment_rate': [0.7, 0.8],\n",
    "                         'Mean_salary': [3200, 3500]})\n",
    "\n",
    "mean_employment_rate_unequal = (0.7 + 0.8) / 2 # This is not the same as (100*0.7 + 250 * 0.8)/ (100+250)\n",
    "\n",
    "# As there are more students for hons. It has a greater effect on the mean than non-hons.\n",
    "print('mean_employment_rate_unequal:', str(mean_employment_rate_unequal))\n",
    "print('Actual employment_rate:', str((100*0.7 + 250 * 0.8)/ (100+250)))\n",
    "\n",
    "# As we only have information about the number of students surveyed but not the sample size for hons vs non-hons for NUS. \n",
    "# We cannot simply use the mean of the two groups "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "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>university</th>\n",
       "      <th>school</th>\n",
       "      <th>degree</th>\n",
       "      <th>employment_rate_overall</th>\n",
       "      <th>employment_rate_ft_perm</th>\n",
       "      <th>basic_monthly_mean</th>\n",
       "      <th>basic_monthly_median</th>\n",
       "      <th>gross_monthly_mean</th>\n",
       "      <th>gross_monthly_median</th>\n",
       "      <th>gross_mthly_25_percentile</th>\n",
       "      <th>gross_mthly_75_percentile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2013</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Accountancy (Hons)</td>\n",
       "      <td>97.1</td>\n",
       "      <td>95.7</td>\n",
       "      <td>2850</td>\n",
       "      <td>2700</td>\n",
       "      <td>2938</td>\n",
       "      <td>2700</td>\n",
       "      <td>2700</td>\n",
       "      <td>2900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2013</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Business (Hons)</td>\n",
       "      <td>90.9</td>\n",
       "      <td>85.7</td>\n",
       "      <td>3053</td>\n",
       "      <td>3000</td>\n",
       "      <td>3214</td>\n",
       "      <td>3000</td>\n",
       "      <td>2700</td>\n",
       "      <td>3500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>2014</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Accountancy (Hons)</td>\n",
       "      <td>96.6</td>\n",
       "      <td>96.1</td>\n",
       "      <td>2981</td>\n",
       "      <td>2800</td>\n",
       "      <td>3003</td>\n",
       "      <td>2840</td>\n",
       "      <td>2700</td>\n",
       "      <td>3000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>87</th>\n",
       "      <td>2014</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Business (Hons)</td>\n",
       "      <td>89.6</td>\n",
       "      <td>80.2</td>\n",
       "      <td>3231</td>\n",
       "      <td>3000</td>\n",
       "      <td>3318</td>\n",
       "      <td>3100</td>\n",
       "      <td>2800</td>\n",
       "      <td>3600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>195</th>\n",
       "      <td>2015</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Accountancy (Hons)</td>\n",
       "      <td>97.3</td>\n",
       "      <td>96.5</td>\n",
       "      <td>3182</td>\n",
       "      <td>2850</td>\n",
       "      <td>3240</td>\n",
       "      <td>2875</td>\n",
       "      <td>2850</td>\n",
       "      <td>3125</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>196</th>\n",
       "      <td>2015</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Business (Hons)</td>\n",
       "      <td>92.2</td>\n",
       "      <td>87.6</td>\n",
       "      <td>3343</td>\n",
       "      <td>3100</td>\n",
       "      <td>3502</td>\n",
       "      <td>3200</td>\n",
       "      <td>2900</td>\n",
       "      <td>3700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>317</th>\n",
       "      <td>2016</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Accountancy (Hons)</td>\n",
       "      <td>97.3</td>\n",
       "      <td>95.1</td>\n",
       "      <td>3027</td>\n",
       "      <td>2900</td>\n",
       "      <td>3063</td>\n",
       "      <td>2900</td>\n",
       "      <td>2900</td>\n",
       "      <td>3100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>319</th>\n",
       "      <td>2016</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Business (Hons)</td>\n",
       "      <td>89.6</td>\n",
       "      <td>77.1</td>\n",
       "      <td>3348</td>\n",
       "      <td>3200</td>\n",
       "      <td>3444</td>\n",
       "      <td>3300</td>\n",
       "      <td>3000</td>\n",
       "      <td>3800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>438</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Accountancy (Hons)</td>\n",
       "      <td>97.1</td>\n",
       "      <td>94.5</td>\n",
       "      <td>3121</td>\n",
       "      <td>3000</td>\n",
       "      <td>3166</td>\n",
       "      <td>3000</td>\n",
       "      <td>2900</td>\n",
       "      <td>3100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>440</th>\n",
       "      <td>2017</td>\n",
       "      <td>NTU</td>\n",
       "      <td>College of Business (Nanyang Business School)</td>\n",
       "      <td>Bachelor of Business (Hons)</td>\n",
       "      <td>93.0</td>\n",
       "      <td>80.8</td>\n",
       "      <td>3530</td>\n",
       "      <td>3300</td>\n",
       "      <td>3643</td>\n",
       "      <td>3300</td>\n",
       "      <td>3000</td>\n",
       "      <td>3800</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     year university                                         school  \\\n",
       "1    2013        NTU  College of Business (Nanyang Business School)   \n",
       "2    2013        NTU  College of Business (Nanyang Business School)   \n",
       "86   2014        NTU  College of Business (Nanyang Business School)   \n",
       "87   2014        NTU  College of Business (Nanyang Business School)   \n",
       "195  2015        NTU  College of Business (Nanyang Business School)   \n",
       "196  2015        NTU  College of Business (Nanyang Business School)   \n",
       "317  2016        NTU  College of Business (Nanyang Business School)   \n",
       "319  2016        NTU  College of Business (Nanyang Business School)   \n",
       "438  2017        NTU  College of Business (Nanyang Business School)   \n",
       "440  2017        NTU  College of Business (Nanyang Business School)   \n",
       "\n",
       "                             degree  employment_rate_overall  \\\n",
       "1    Bachelor of Accountancy (Hons)                     97.1   \n",
       "2       Bachelor of Business (Hons)                     90.9   \n",
       "86   Bachelor of Accountancy (Hons)                     96.6   \n",
       "87      Bachelor of Business (Hons)                     89.6   \n",
       "195  Bachelor of Accountancy (Hons)                     97.3   \n",
       "196     Bachelor of Business (Hons)                     92.2   \n",
       "317  Bachelor of Accountancy (Hons)                     97.3   \n",
       "319     Bachelor of Business (Hons)                     89.6   \n",
       "438  Bachelor of Accountancy (Hons)                     97.1   \n",
       "440     Bachelor of Business (Hons)                     93.0   \n",
       "\n",
       "     employment_rate_ft_perm  basic_monthly_mean  basic_monthly_median  \\\n",
       "1                       95.7                2850                  2700   \n",
       "2                       85.7                3053                  3000   \n",
       "86                      96.1                2981                  2800   \n",
       "87                      80.2                3231                  3000   \n",
       "195                     96.5                3182                  2850   \n",
       "196                     87.6                3343                  3100   \n",
       "317                     95.1                3027                  2900   \n",
       "319                     77.1                3348                  3200   \n",
       "438                     94.5                3121                  3000   \n",
       "440                     80.8                3530                  3300   \n",
       "\n",
       "     gross_monthly_mean  gross_monthly_median  gross_mthly_25_percentile  \\\n",
       "1                  2938                  2700                       2700   \n",
       "2                  3214                  3000                       2700   \n",
       "86                 3003                  2840                       2700   \n",
       "87                 3318                  3100                       2800   \n",
       "195                3240                  2875                       2850   \n",
       "196                3502                  3200                       2900   \n",
       "317                3063                  2900                       2900   \n",
       "319                3444                  3300                       3000   \n",
       "438                3166                  3000                       2900   \n",
       "440                3643                  3300                       3000   \n",
       "\n",
       "     gross_mthly_75_percentile  \n",
       "1                         2900  \n",
       "2                         3500  \n",
       "86                        3000  \n",
       "87                        3600  \n",
       "195                       3125  \n",
       "196                       3700  \n",
       "317                       3100  \n",
       "319                       3800  \n",
       "438                       3100  \n",
       "440                       3800  "
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Split dataset into university level df_'university' and subset out only business school, single degree, ensure consistent naming.\n",
    "\n",
    "# NTU\n",
    "df_ntu = df[(df['university'] == 'NTU') & (df['school'] == 'College of Business (Nanyang Business School)')]\n",
    "\n",
    "# By removing the word 'and', we effectively removed all double degree information since the naming convention for double degree follows '1st degree and 2nd degree'\n",
    "# ~ represents NOT \n",
    "df_ntu = df_ntu[~df_ntu['degree'].str.contains('and')]\n",
    "\n",
    "# Rename older degrees to follow the current naming convention for ease of understanding\n",
    "df_ntu.loc[df_ntu['degree'].str.contains('Accountancy'), ['degree']] = 'Bachelor of Accountancy (Hons)'\n",
    "\n",
    "df_ntu.loc[df_ntu['degree'].str.contains('Business'), ['degree']] = 'Bachelor of Business (Hons)'\n",
    "\n",
    "df_ntu"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "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>university</th>\n",
       "      <th>school</th>\n",
       "      <th>degree</th>\n",
       "      <th>employment_rate_overall</th>\n",
       "      <th>employment_rate_ft_perm</th>\n",
       "      <th>basic_monthly_mean</th>\n",
       "      <th>basic_monthly_median</th>\n",
       "      <th>gross_monthly_mean</th>\n",
       "      <th>gross_monthly_median</th>\n",
       "      <th>gross_mthly_25_percentile</th>\n",
       "      <th>gross_mthly_75_percentile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration</td>\n",
       "      <td>88.7</td>\n",
       "      <td>81.7</td>\n",
       "      <td>2960</td>\n",
       "      <td>2868</td>\n",
       "      <td>3062</td>\n",
       "      <td>3000</td>\n",
       "      <td>2700</td>\n",
       "      <td>3370</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Hons)</td>\n",
       "      <td>94.6</td>\n",
       "      <td>90.3</td>\n",
       "      <td>3404</td>\n",
       "      <td>3050</td>\n",
       "      <td>3512</td>\n",
       "      <td>3200</td>\n",
       "      <td>2800</td>\n",
       "      <td>3800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountancy)</td>\n",
       "      <td>93.3</td>\n",
       "      <td>88.9</td>\n",
       "      <td>2740</td>\n",
       "      <td>2700</td>\n",
       "      <td>2922</td>\n",
       "      <td>2700</td>\n",
       "      <td>2700</td>\n",
       "      <td>3018</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>2013</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountan...</td>\n",
       "      <td>97.2</td>\n",
       "      <td>94.4</td>\n",
       "      <td>3065</td>\n",
       "      <td>2800</td>\n",
       "      <td>3143</td>\n",
       "      <td>2800</td>\n",
       "      <td>2700</td>\n",
       "      <td>3200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>122</th>\n",
       "      <td>2014</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration</td>\n",
       "      <td>87.9</td>\n",
       "      <td>83.6</td>\n",
       "      <td>3234</td>\n",
       "      <td>3000</td>\n",
       "      <td>3367</td>\n",
       "      <td>3000</td>\n",
       "      <td>2800</td>\n",
       "      <td>3300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>2014</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Hons)</td>\n",
       "      <td>94.4</td>\n",
       "      <td>90.6</td>\n",
       "      <td>3979</td>\n",
       "      <td>3500</td>\n",
       "      <td>4090</td>\n",
       "      <td>3520</td>\n",
       "      <td>3200</td>\n",
       "      <td>4300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>2014</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountancy)</td>\n",
       "      <td>93.9</td>\n",
       "      <td>89.8</td>\n",
       "      <td>2836</td>\n",
       "      <td>2800</td>\n",
       "      <td>2989</td>\n",
       "      <td>2838</td>\n",
       "      <td>2700</td>\n",
       "      <td>3000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>125</th>\n",
       "      <td>2014</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountan...</td>\n",
       "      <td>100.0</td>\n",
       "      <td>96.3</td>\n",
       "      <td>3350</td>\n",
       "      <td>2850</td>\n",
       "      <td>3407</td>\n",
       "      <td>2912</td>\n",
       "      <td>2800</td>\n",
       "      <td>3614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>236</th>\n",
       "      <td>2015</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration</td>\n",
       "      <td>88.2</td>\n",
       "      <td>80.3</td>\n",
       "      <td>3164</td>\n",
       "      <td>3000</td>\n",
       "      <td>3341</td>\n",
       "      <td>3000</td>\n",
       "      <td>2800</td>\n",
       "      <td>3700</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>237</th>\n",
       "      <td>2015</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Hons)</td>\n",
       "      <td>96.0</td>\n",
       "      <td>93.1</td>\n",
       "      <td>4173</td>\n",
       "      <td>3750</td>\n",
       "      <td>4326</td>\n",
       "      <td>3800</td>\n",
       "      <td>3300</td>\n",
       "      <td>4600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>238</th>\n",
       "      <td>2015</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountancy)</td>\n",
       "      <td>97.6</td>\n",
       "      <td>97.6</td>\n",
       "      <td>2817</td>\n",
       "      <td>2850</td>\n",
       "      <td>2983</td>\n",
       "      <td>2850</td>\n",
       "      <td>2700</td>\n",
       "      <td>2960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>239</th>\n",
       "      <td>2015</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountan...</td>\n",
       "      <td>98.6</td>\n",
       "      <td>98.6</td>\n",
       "      <td>3547</td>\n",
       "      <td>3000</td>\n",
       "      <td>3587</td>\n",
       "      <td>3000</td>\n",
       "      <td>2850</td>\n",
       "      <td>3634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>377</th>\n",
       "      <td>2016</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration</td>\n",
       "      <td>93.6</td>\n",
       "      <td>80.7</td>\n",
       "      <td>3232</td>\n",
       "      <td>3000</td>\n",
       "      <td>3282</td>\n",
       "      <td>3000</td>\n",
       "      <td>2837</td>\n",
       "      <td>3500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>378</th>\n",
       "      <td>2016</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Hons)</td>\n",
       "      <td>96.4</td>\n",
       "      <td>92.8</td>\n",
       "      <td>4126</td>\n",
       "      <td>3600</td>\n",
       "      <td>4207</td>\n",
       "      <td>3620</td>\n",
       "      <td>3258</td>\n",
       "      <td>4500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>379</th>\n",
       "      <td>2016</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountancy)</td>\n",
       "      <td>88.6</td>\n",
       "      <td>86.4</td>\n",
       "      <td>2977</td>\n",
       "      <td>2900</td>\n",
       "      <td>3020</td>\n",
       "      <td>2963</td>\n",
       "      <td>2800</td>\n",
       "      <td>3100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>380</th>\n",
       "      <td>2016</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountan...</td>\n",
       "      <td>99.0</td>\n",
       "      <td>99.0</td>\n",
       "      <td>3435</td>\n",
       "      <td>3000</td>\n",
       "      <td>3465</td>\n",
       "      <td>3075</td>\n",
       "      <td>2900</td>\n",
       "      <td>3800</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>501</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration</td>\n",
       "      <td>94.9</td>\n",
       "      <td>87.3</td>\n",
       "      <td>3770</td>\n",
       "      <td>3400</td>\n",
       "      <td>4031</td>\n",
       "      <td>3500</td>\n",
       "      <td>3000</td>\n",
       "      <td>4333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>502</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Hons)</td>\n",
       "      <td>98.2</td>\n",
       "      <td>93.6</td>\n",
       "      <td>4272</td>\n",
       "      <td>3725</td>\n",
       "      <td>4388</td>\n",
       "      <td>3800</td>\n",
       "      <td>3450</td>\n",
       "      <td>4782</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>503</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountancy)</td>\n",
       "      <td>97.2</td>\n",
       "      <td>94.4</td>\n",
       "      <td>3396</td>\n",
       "      <td>3000</td>\n",
       "      <td>3473</td>\n",
       "      <td>3025</td>\n",
       "      <td>3000</td>\n",
       "      <td>3600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>504</th>\n",
       "      <td>2017</td>\n",
       "      <td>NUS</td>\n",
       "      <td>NUS Business School</td>\n",
       "      <td>Bachelor of Business Administration (Accountan...</td>\n",
       "      <td>100.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>3689</td>\n",
       "      <td>3100</td>\n",
       "      <td>3744</td>\n",
       "      <td>3125</td>\n",
       "      <td>3000</td>\n",
       "      <td>4023</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     year university               school  \\\n",
       "37   2013        NUS  NUS Business School   \n",
       "38   2013        NUS  NUS Business School   \n",
       "39   2013        NUS  NUS Business School   \n",
       "40   2013        NUS  NUS Business School   \n",
       "122  2014        NUS  NUS Business School   \n",
       "123  2014        NUS  NUS Business School   \n",
       "124  2014        NUS  NUS Business School   \n",
       "125  2014        NUS  NUS Business School   \n",
       "236  2015        NUS  NUS Business School   \n",
       "237  2015        NUS  NUS Business School   \n",
       "238  2015        NUS  NUS Business School   \n",
       "239  2015        NUS  NUS Business School   \n",
       "377  2016        NUS  NUS Business School   \n",
       "378  2016        NUS  NUS Business School   \n",
       "379  2016        NUS  NUS Business School   \n",
       "380  2016        NUS  NUS Business School   \n",
       "501  2017        NUS  NUS Business School   \n",
       "502  2017        NUS  NUS Business School   \n",
       "503  2017        NUS  NUS Business School   \n",
       "504  2017        NUS  NUS Business School   \n",
       "\n",
       "                                                degree  \\\n",
       "37                 Bachelor of Business Administration   \n",
       "38          Bachelor of Business Administration (Hons)   \n",
       "39   Bachelor of Business Administration (Accountancy)   \n",
       "40   Bachelor of Business Administration (Accountan...   \n",
       "122                Bachelor of Business Administration   \n",
       "123         Bachelor of Business Administration (Hons)   \n",
       "124  Bachelor of Business Administration (Accountancy)   \n",
       "125  Bachelor of Business Administration (Accountan...   \n",
       "236                Bachelor of Business Administration   \n",
       "237         Bachelor of Business Administration (Hons)   \n",
       "238  Bachelor of Business Administration (Accountancy)   \n",
       "239  Bachelor of Business Administration (Accountan...   \n",
       "377                Bachelor of Business Administration   \n",
       "378         Bachelor of Business Administration (Hons)   \n",
       "379  Bachelor of Business Administration (Accountancy)   \n",
       "380  Bachelor of Business Administration (Accountan...   \n",
       "501                Bachelor of Business Administration   \n",
       "502         Bachelor of Business Administration (Hons)   \n",
       "503  Bachelor of Business Administration (Accountancy)   \n",
       "504  Bachelor of Business Administration (Accountan...   \n",
       "\n",
       "     employment_rate_overall  employment_rate_ft_perm  basic_monthly_mean  \\\n",
       "37                      88.7                     81.7                2960   \n",
       "38                      94.6                     90.3                3404   \n",
       "39                      93.3                     88.9                2740   \n",
       "40                      97.2                     94.4                3065   \n",
       "122                     87.9                     83.6                3234   \n",
       "123                     94.4                     90.6                3979   \n",
       "124                     93.9                     89.8                2836   \n",
       "125                    100.0                     96.3                3350   \n",
       "236                     88.2                     80.3                3164   \n",
       "237                     96.0                     93.1                4173   \n",
       "238                     97.6                     97.6                2817   \n",
       "239                     98.6                     98.6                3547   \n",
       "377                     93.6                     80.7                3232   \n",
       "378                     96.4                     92.8                4126   \n",
       "379                     88.6                     86.4                2977   \n",
       "380                     99.0                     99.0                3435   \n",
       "501                     94.9                     87.3                3770   \n",
       "502                     98.2                     93.6                4272   \n",
       "503                     97.2                     94.4                3396   \n",
       "504                    100.0                    100.0                3689   \n",
       "\n",
       "     basic_monthly_median  gross_monthly_mean  gross_monthly_median  \\\n",
       "37                   2868                3062                  3000   \n",
       "38                   3050                3512                  3200   \n",
       "39                   2700                2922                  2700   \n",
       "40                   2800                3143                  2800   \n",
       "122                  3000                3367                  3000   \n",
       "123                  3500                4090                  3520   \n",
       "124                  2800                2989                  2838   \n",
       "125                  2850                3407                  2912   \n",
       "236                  3000                3341                  3000   \n",
       "237                  3750                4326                  3800   \n",
       "238                  2850                2983                  2850   \n",
       "239                  3000                3587                  3000   \n",
       "377                  3000                3282                  3000   \n",
       "378                  3600                4207                  3620   \n",
       "379                  2900                3020                  2963   \n",
       "380                  3000                3465                  3075   \n",
       "501                  3400                4031                  3500   \n",
       "502                  3725                4388                  3800   \n",
       "503                  3000                3473                  3025   \n",
       "504                  3100                3744                  3125   \n",
       "\n",
       "     gross_mthly_25_percentile  gross_mthly_75_percentile  \n",
       "37                        2700                       3370  \n",
       "38                        2800                       3800  \n",
       "39                        2700                       3018  \n",
       "40                        2700                       3200  \n",
       "122                       2800                       3300  \n",
       "123                       3200                       4300  \n",
       "124                       2700                       3000  \n",
       "125                       2800                       3614  \n",
       "236                       2800                       3700  \n",
       "237                       3300                       4600  \n",
       "238                       2700                       2960  \n",
       "239                       2850                       3634  \n",
       "377                       2837                       3500  \n",
       "378                       3258                       4500  \n",
       "379                       2800                       3100  \n",
       "380                       2900                       3800  \n",
       "501                       3000                       4333  \n",
       "502                       3450                       4782  \n",
       "503                       3000                       3600  \n",
       "504                       3000                       4023  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# NUS\n",
    "# This dataset does not contain NUS double degree information. Hence, we simply need to subset business school\n",
    "df_nus = df[(df['university'] == 'NUS') & (df['school'] == 'NUS Business School')]\n",
    "\n",
    "df_nus"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "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>university</th>\n",
       "      <th>school</th>\n",
       "      <th>degree</th>\n",
       "      <th>employment_rate_overall</th>\n",
       "      <th>employment_rate_ft_perm</th>\n",
       "      <th>basic_monthly_mean</th>\n",
       "      <th>basic_monthly_median</th>\n",
       "      <th>gross_monthly_mean</th>\n",
       "      <th>gross_monthly_median</th>\n",
       "      <th>gross_mthly_25_percentile</th>\n",
       "      <th>gross_mthly_75_percentile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>2013</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-years programme) *</td>\n",
       "      <td>Accountancy</td>\n",
       "      <td>94.2</td>\n",
       "      <td>93.2</td>\n",
       "      <td>3231</td>\n",
       "      <td>2800</td>\n",
       "      <td>3342</td>\n",
       "      <td>2825</td>\n",
       "      <td>2700</td>\n",
       "      <td>3430</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>74</th>\n",
       "      <td>2013</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-years programme) *</td>\n",
       "      <td>Accountancy Cum Laude and above</td>\n",
       "      <td>95.2</td>\n",
       "      <td>95.2</td>\n",
       "      <td>3580</td>\n",
       "      <td>2900</td>\n",
       "      <td>3634</td>\n",
       "      <td>3000</td>\n",
       "      <td>2700</td>\n",
       "      <td>3750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>2013</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-years programme) *</td>\n",
       "      <td>Business Management</td>\n",
       "      <td>93.2</td>\n",
       "      <td>87.3</td>\n",
       "      <td>3344</td>\n",
       "      <td>3000</td>\n",
       "      <td>3520</td>\n",
       "      <td>3200</td>\n",
       "      <td>2800</td>\n",
       "      <td>3750</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>2013</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-years programme) *</td>\n",
       "      <td>Business Management Cum Laude and above</td>\n",
       "      <td>95.9</td>\n",
       "      <td>93.8</td>\n",
       "      <td>3825</td>\n",
       "      <td>3500</td>\n",
       "      <td>3975</td>\n",
       "      <td>3580</td>\n",
       "      <td>3000</td>\n",
       "      <td>4100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>158</th>\n",
       "      <td>2014</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-year programme) *</td>\n",
       "      <td>Accountancy</td>\n",
       "      <td>93.6</td>\n",
       "      <td>91.4</td>\n",
       "      <td>3375</td>\n",
       "      <td>2850</td>\n",
       "      <td>3420</td>\n",
       "      <td>2875</td>\n",
       "      <td>2722</td>\n",
       "      <td>3614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>159</th>\n",
       "      <td>2014</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-year programme) *</td>\n",
       "      <td>Accountancy Cum Laude and above</td>\n",
       "      <td>97.4</td>\n",
       "      <td>96.1</td>\n",
       "      <td>3871</td>\n",
       "      <td>3450</td>\n",
       "      <td>3903</td>\n",
       "      <td>3500</td>\n",
       "      <td>2875</td>\n",
       "      <td>4300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>160</th>\n",
       "      <td>2014</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-year programme) *</td>\n",
       "      <td>Business Management</td>\n",
       "      <td>90.9</td>\n",
       "      <td>86.4</td>\n",
       "      <td>3630</td>\n",
       "      <td>3250</td>\n",
       "      <td>3752</td>\n",
       "      <td>3333</td>\n",
       "      <td>3000</td>\n",
       "      <td>4000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161</th>\n",
       "      <td>2014</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-year programme) *</td>\n",
       "      <td>Business Management Cum Laude and above</td>\n",
       "      <td>94.5</td>\n",
       "      <td>94.0</td>\n",
       "      <td>4019</td>\n",
       "      <td>3500</td>\n",
       "      <td>4116</td>\n",
       "      <td>3600</td>\n",
       "      <td>3200</td>\n",
       "      <td>4500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>273</th>\n",
       "      <td>2015</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-year programme) *</td>\n",
       "      <td>Accountancy</td>\n",
       "      <td>97.3</td>\n",
       "      <td>93.0</td>\n",
       "      <td>3287</td>\n",
       "      <td>2900</td>\n",
       "      <td>3427</td>\n",
       "      <td>3000</td>\n",
       "      <td>2850</td>\n",
       "      <td>3500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>274</th>\n",
       "      <td>2015</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-year programme) *</td>\n",
       "      <td>Accountancy Cum Laude and above</td>\n",
       "      <td>97.7</td>\n",
       "      <td>97.7</td>\n",
       "      <td>3597</td>\n",
       "      <td>3100</td>\n",
       "      <td>3701</td>\n",
       "      <td>3200</td>\n",
       "      <td>2850</td>\n",
       "      <td>4110</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>275</th>\n",
       "      <td>2015</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-year programme) *</td>\n",
       "      <td>Business Management</td>\n",
       "      <td>93.9</td>\n",
       "      <td>88.2</td>\n",
       "      <td>3513</td>\n",
       "      <td>3200</td>\n",
       "      <td>3611</td>\n",
       "      <td>3300</td>\n",
       "      <td>3000</td>\n",
       "      <td>4000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>276</th>\n",
       "      <td>2015</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-year programme) *</td>\n",
       "      <td>Business Management Cum Laude and above</td>\n",
       "      <td>98.3</td>\n",
       "      <td>93.3</td>\n",
       "      <td>4050</td>\n",
       "      <td>3800</td>\n",
       "      <td>4130</td>\n",
       "      <td>3868</td>\n",
       "      <td>3200</td>\n",
       "      <td>4600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>394</th>\n",
       "      <td>2016</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-year programme) *</td>\n",
       "      <td>Accountancy</td>\n",
       "      <td>98.5</td>\n",
       "      <td>96.0</td>\n",
       "      <td>3390</td>\n",
       "      <td>3000</td>\n",
       "      <td>3455</td>\n",
       "      <td>3000</td>\n",
       "      <td>2900</td>\n",
       "      <td>3600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>395</th>\n",
       "      <td>2016</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-year programme) *</td>\n",
       "      <td>Accountancy Cum Laude and above</td>\n",
       "      <td>100.0</td>\n",
       "      <td>100.0</td>\n",
       "      <td>3827</td>\n",
       "      <td>3500</td>\n",
       "      <td>3943</td>\n",
       "      <td>3600</td>\n",
       "      <td>3000</td>\n",
       "      <td>4583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>396</th>\n",
       "      <td>2016</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-year programme) *</td>\n",
       "      <td>Business Management</td>\n",
       "      <td>92.4</td>\n",
       "      <td>84.6</td>\n",
       "      <td>3727</td>\n",
       "      <td>3400</td>\n",
       "      <td>3831</td>\n",
       "      <td>3500</td>\n",
       "      <td>3000</td>\n",
       "      <td>4000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>397</th>\n",
       "      <td>2016</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-year programme) *</td>\n",
       "      <td>Business Management Cum Laude and above</td>\n",
       "      <td>96.1</td>\n",
       "      <td>92.1</td>\n",
       "      <td>4140</td>\n",
       "      <td>3550</td>\n",
       "      <td>4234</td>\n",
       "      <td>3660</td>\n",
       "      <td>3250</td>\n",
       "      <td>4583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>524</th>\n",
       "      <td>2017</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-years programme) *</td>\n",
       "      <td>Accountancy</td>\n",
       "      <td>98.0</td>\n",
       "      <td>96.6</td>\n",
       "      <td>3569</td>\n",
       "      <td>3000</td>\n",
       "      <td>3646</td>\n",
       "      <td>3000</td>\n",
       "      <td>2920</td>\n",
       "      <td>3600</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>525</th>\n",
       "      <td>2017</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Accountancy (4-years programme) *</td>\n",
       "      <td>Accountancy Cum Laude and above</td>\n",
       "      <td>97.8</td>\n",
       "      <td>97.8</td>\n",
       "      <td>4037</td>\n",
       "      <td>3200</td>\n",
       "      <td>4089</td>\n",
       "      <td>3200</td>\n",
       "      <td>3000</td>\n",
       "      <td>4645</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>526</th>\n",
       "      <td>2017</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-years programme) *</td>\n",
       "      <td>Business Management</td>\n",
       "      <td>93.3</td>\n",
       "      <td>85.7</td>\n",
       "      <td>3862</td>\n",
       "      <td>3475</td>\n",
       "      <td>3995</td>\n",
       "      <td>3500</td>\n",
       "      <td>3050</td>\n",
       "      <td>4362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>527</th>\n",
       "      <td>2017</td>\n",
       "      <td>SMU</td>\n",
       "      <td>School of Business (4-years programme) *</td>\n",
       "      <td>Business Management Cum Laude and above</td>\n",
       "      <td>95.8</td>\n",
       "      <td>92.0</td>\n",
       "      <td>4364</td>\n",
       "      <td>3880</td>\n",
       "      <td>4504</td>\n",
       "      <td>4000</td>\n",
       "      <td>3300</td>\n",
       "      <td>5000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     year university                                       school  \\\n",
       "73   2013        SMU  School of Accountancy (4-years programme) *   \n",
       "74   2013        SMU  School of Accountancy (4-years programme) *   \n",
       "75   2013        SMU     School of Business (4-years programme) *   \n",
       "76   2013        SMU     School of Business (4-years programme) *   \n",
       "158  2014        SMU   School of Accountancy (4-year programme) *   \n",
       "159  2014        SMU   School of Accountancy (4-year programme) *   \n",
       "160  2014        SMU      School of Business (4-year programme) *   \n",
       "161  2014        SMU      School of Business (4-year programme) *   \n",
       "273  2015        SMU   School of Accountancy (4-year programme) *   \n",
       "274  2015        SMU   School of Accountancy (4-year programme) *   \n",
       "275  2015        SMU      School of Business (4-year programme) *   \n",
       "276  2015        SMU      School of Business (4-year programme) *   \n",
       "394  2016        SMU   School of Accountancy (4-year programme) *   \n",
       "395  2016        SMU   School of Accountancy (4-year programme) *   \n",
       "396  2016        SMU      School of Business (4-year programme) *   \n",
       "397  2016        SMU      School of Business (4-year programme) *   \n",
       "524  2017        SMU  School of Accountancy (4-years programme) *   \n",
       "525  2017        SMU  School of Accountancy (4-years programme) *   \n",
       "526  2017        SMU     School of Business (4-years programme) *   \n",
       "527  2017        SMU     School of Business (4-years programme) *   \n",
       "\n",
       "                                      degree  employment_rate_overall  \\\n",
       "73                               Accountancy                     94.2   \n",
       "74           Accountancy Cum Laude and above                     95.2   \n",
       "75                       Business Management                     93.2   \n",
       "76   Business Management Cum Laude and above                     95.9   \n",
       "158                              Accountancy                     93.6   \n",
       "159          Accountancy Cum Laude and above                     97.4   \n",
       "160                      Business Management                     90.9   \n",
       "161  Business Management Cum Laude and above                     94.5   \n",
       "273                              Accountancy                     97.3   \n",
       "274          Accountancy Cum Laude and above                     97.7   \n",
       "275                      Business Management                     93.9   \n",
       "276  Business Management Cum Laude and above                     98.3   \n",
       "394                              Accountancy                     98.5   \n",
       "395          Accountancy Cum Laude and above                    100.0   \n",
       "396                      Business Management                     92.4   \n",
       "397  Business Management Cum Laude and above                     96.1   \n",
       "524                              Accountancy                     98.0   \n",
       "525          Accountancy Cum Laude and above                     97.8   \n",
       "526                      Business Management                     93.3   \n",
       "527  Business Management Cum Laude and above                     95.8   \n",
       "\n",
       "     employment_rate_ft_perm  basic_monthly_mean  basic_monthly_median  \\\n",
       "73                      93.2                3231                  2800   \n",
       "74                      95.2                3580                  2900   \n",
       "75                      87.3                3344                  3000   \n",
       "76                      93.8                3825                  3500   \n",
       "158                     91.4                3375                  2850   \n",
       "159                     96.1                3871                  3450   \n",
       "160                     86.4                3630                  3250   \n",
       "161                     94.0                4019                  3500   \n",
       "273                     93.0                3287                  2900   \n",
       "274                     97.7                3597                  3100   \n",
       "275                     88.2                3513                  3200   \n",
       "276                     93.3                4050                  3800   \n",
       "394                     96.0                3390                  3000   \n",
       "395                    100.0                3827                  3500   \n",
       "396                     84.6                3727                  3400   \n",
       "397                     92.1                4140                  3550   \n",
       "524                     96.6                3569                  3000   \n",
       "525                     97.8                4037                  3200   \n",
       "526                     85.7                3862                  3475   \n",
       "527                     92.0                4364                  3880   \n",
       "\n",
       "     gross_monthly_mean  gross_monthly_median  gross_mthly_25_percentile  \\\n",
       "73                 3342                  2825                       2700   \n",
       "74                 3634                  3000                       2700   \n",
       "75                 3520                  3200                       2800   \n",
       "76                 3975                  3580                       3000   \n",
       "158                3420                  2875                       2722   \n",
       "159                3903                  3500                       2875   \n",
       "160                3752                  3333                       3000   \n",
       "161                4116                  3600                       3200   \n",
       "273                3427                  3000                       2850   \n",
       "274                3701                  3200                       2850   \n",
       "275                3611                  3300                       3000   \n",
       "276                4130                  3868                       3200   \n",
       "394                3455                  3000                       2900   \n",
       "395                3943                  3600                       3000   \n",
       "396                3831                  3500                       3000   \n",
       "397                4234                  3660                       3250   \n",
       "524                3646                  3000                       2920   \n",
       "525                4089                  3200                       3000   \n",
       "526                3995                  3500                       3050   \n",
       "527                4504                  4000                       3300   \n",
       "\n",
       "     gross_mthly_75_percentile  \n",
       "73                        3430  \n",
       "74                        3750  \n",
       "75                        3750  \n",
       "76                        4100  \n",
       "158                       3614  \n",
       "159                       4300  \n",
       "160                       4000  \n",
       "161                       4500  \n",
       "273                       3500  \n",
       "274                       4110  \n",
       "275                       4000  \n",
       "276                       4600  \n",
       "394                       3600  \n",
       "395                       4583  \n",
       "396                       4000  \n",
       "397                       4583  \n",
       "524                       3600  \n",
       "525                       4645  \n",
       "526                       4362  \n",
       "527                       5000  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# SMU\n",
    "\n",
    "# While, SMU classifies its Accountancy and Business degrees as different schools. It is ok because we are comparing internally\n",
    "# As SMU has information on both the general cohort ie any award and also Cum Laude and above. We will keep both for the time being and observe the trend later\n",
    "df_smu = df[df['university'] == 'SMU']\n",
    "\n",
    "# Having a boolean inside the square bracket of a data frame enables Python to return all of the true cases\n",
    "df_smu = df_smu[df_smu['school'].str.contains('Accountancy|Business')]\n",
    "\n",
    "# Rename older degrees to follow the current naming convention for ease of understanding\n",
    "# 2013 was (4-years programme) while 2014 was (4-year programme). Hence, need to change both years.\n",
    "\n",
    "for i in df_smu['degree']:\n",
    "    if 'Accountancy' in i.split() and 'Laude' in i.split():\n",
    "        df_smu['degree'] = df_smu['degree'].replace(i, 'Accountancy Cum Laude and above')\n",
    "    elif 'Business' in i.split() and 'Laude' in i.split():\n",
    "        df_smu['degree'] = df_smu['degree'].replace(i, 'Business Management Cum Laude and above')        \n",
    "    elif 'Accountancy' in i.split():\n",
    "        df_smu['degree'] = df_smu['degree'].replace(i, 'Accountancy')\n",
    "    elif 'Business' in i.split():\n",
    "        df_smu['degree'] = df_smu['degree'].replace(i, 'Business Management')\n",
    "\n",
    "df_smu"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "scrolled": true
   },
   "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>university</th>\n",
       "      <th>school</th>\n",
       "      <th>degree</th>\n",
       "      <th>employment_rate_overall</th>\n",
       "      <th>employment_rate_ft_perm</th>\n",
       "      <th>basic_monthly_mean</th>\n",
       "      <th>basic_monthly_median</th>\n",
       "      <th>gross_monthly_mean</th>\n",
       "      <th>gross_monthly_median</th>\n",
       "      <th>gross_mthly_25_percentile</th>\n",
       "      <th>gross_mthly_75_percentile</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>564</th>\n",
       "      <td>2017</td>\n",
       "      <td>SIT</td>\n",
       "      <td>Singapore Institute of Technology</td>\n",
       "      <td>Bachelor of Accountancy with Honours</td>\n",
       "      <td>94.1</td>\n",
       "      <td>92.1</td>\n",
       "      <td>3073</td>\n",
       "      <td>3050</td>\n",
       "      <td>3113</td>\n",
       "      <td>3088</td>\n",
       "      <td>2975</td>\n",
       "      <td>3250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>565</th>\n",
       "      <td>2017</td>\n",
       "      <td>SIT</td>\n",
       "      <td>Singapore Institute of Technology</td>\n",
       "      <td>Bachelor of Hospitality Business with Honours ^^^</td>\n",
       "      <td>94.7</td>\n",
       "      <td>84.2</td>\n",
       "      <td>2713</td>\n",
       "      <td>2700</td>\n",
       "      <td>2829</td>\n",
       "      <td>2800</td>\n",
       "      <td>2500</td>\n",
       "      <td>3100</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     year university                             school  \\\n",
       "564  2017        SIT  Singapore Institute of Technology   \n",
       "565  2017        SIT  Singapore Institute of Technology   \n",
       "\n",
       "                                                degree  \\\n",
       "564               Bachelor of Accountancy with Honours   \n",
       "565  Bachelor of Hospitality Business with Honours ^^^   \n",
       "\n",
       "     employment_rate_overall  employment_rate_ft_perm  basic_monthly_mean  \\\n",
       "564                     94.1                     92.1                3073   \n",
       "565                     94.7                     84.2                2713   \n",
       "\n",
       "     basic_monthly_median  gross_monthly_mean  gross_monthly_median  \\\n",
       "564                  3050                3113                  3088   \n",
       "565                  2700                2829                  2800   \n",
       "\n",
       "     gross_mthly_25_percentile  gross_mthly_75_percentile  \n",
       "564                       2975                       3250  \n",
       "565                       2500                       3100  "
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# SIT\n",
    "\n",
    "# SIT only has one year of data for Accountancy and Business related degrees.\n",
    "# Bachelor of Hospitality Business with Honours has fewer than 30 sample size (Check metadata) \n",
    "df_sit = df[(df['university'] == 'SIT') & (df['school'] == 'Singapore Institute of Technology')]\n",
    "df_sit"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# After ensure naming consistency, we will merge back the dataset of each school less SIT since it has only one year of data\n",
    "# Due to the lack of data for SIT, I have decided to remove from the analysis \n",
    "df_comb = df_ntu.append([df_nus, df_smu])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a generic function that can automatically generate emp rate plots for each university\n",
    "# A_Hons and B_Hons represent Accountancy and Business with Honours respectively\n",
    "# A_Cum Laude and B_Cum Laude represent Accountancy and Business with Cum Laude and above respectively (Exclusive to SMU)\n",
    "# A and B could mean Accountancy and Business without Hons/ Cum Laude depending on the university\n",
    "\n",
    "def gen_emp_rate(arr, unq_deg):\n",
    "    for i in unq_deg:\n",
    "        if '(Hons)' in i.split(): # NTU and NUS have (Hons) cases\n",
    "            if '(Accountancy)' in i.split() or 'Accountancy' in i.split(): # NUS has bracket while NTU does not\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['employment_rate_ft_perm'],\n",
    "                         color = 'red', linestyle = '--', label = 'A_Hons')               \n",
    "            elif 'Business' in i.split():\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['employment_rate_ft_perm'],\n",
    "                         color = 'blue', linestyle = '--', label = 'B_Hons')                \n",
    "        elif 'Laude' in i.split(): # For SMU Cum Laude \n",
    "            if 'Accountancy' in i.split():\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['employment_rate_ft_perm'],\n",
    "                         color = 'red', linestyle = '--', label = 'A_Cum Laude')\n",
    "            elif 'Business' in i.split():\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['employment_rate_ft_perm'],\n",
    "                         color = 'blue', linestyle = '--', label = 'B_Cum Laude')\n",
    "        elif '(Accountancy)' in i.split() or 'Accountancy' in i.split(): # NUS Accountancy has bracket\n",
    "            plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                     arr[arr['degree'] == i]['employment_rate_ft_perm'],\n",
    "                     color = 'red', label = 'A')\n",
    "        elif 'Business' in i.split(): # For NUS Business w/o (Hons) and SMU Business w/o Cum Laude \n",
    "            plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                     arr[arr['degree'] == i]['employment_rate_ft_perm'],\n",
    "                     color = 'blue', label = 'B')\n",
    "    plt.ylabel('employment_rate_ft_perm')\n",
    "    plt.xlabel('year')\n",
    "    plt.xticks(list(arr['year']))\n",
    "    plt.title(str(uni) + ' Employment_rate between 2013 - 2017') # The for loop below ensures that the appropriate title gets called\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a generic function that can automatically generate median salary plot for each university\n",
    "# A_Hons and B_Hons represent Accountancy and Business with Honours respectively\n",
    "# A_Cum Laude and B_Cum Laude represent Accountancy and Business with Cum Laude and above respectively (Exclusive to SMU)\n",
    "# A and B could mean Accountancy and Business without Hons/ Cum Laude depending on the university\n",
    "\n",
    "def gen_median_salary(arr, unq_deg):\n",
    "    for i in unq_deg:\n",
    "        if '(Hons)' in i.split(): # NTU and NUS have (Hons) cases\n",
    "            if '(Accountancy)' in i.split() or 'Accountancy' in i.split(): # NUS has bracket while NTU does not\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['basic_monthly_median'],\n",
    "                         color = 'red', linestyle = '--', label = 'A_Hons')               \n",
    "            elif 'Business' in i.split():\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['basic_monthly_median'],\n",
    "                         color = 'blue', linestyle = '--', label = 'B_Hons')                \n",
    "        elif 'Laude' in i.split(): # For SMU Cum Laude \n",
    "            if 'Accountancy' in i.split():\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['basic_monthly_median'],\n",
    "                         color = 'red', linestyle = '--', label = 'A_Cum Laude')\n",
    "            elif 'Business' in i.split():\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['basic_monthly_median'],\n",
    "                         color = 'blue', linestyle = '--', label = 'B_Cum Laude')\n",
    "        elif '(Accountancy)' in i.split() or 'Accountancy' in i.split(): # NUS Accountancy has bracket\n",
    "            plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                     arr[arr['degree'] == i]['basic_monthly_median'],\n",
    "                     color = 'red', label = 'A')\n",
    "        elif 'Business' in i.split(): # For NUS Business w/o (Hons) and SMU Business w/o Cum Laude \n",
    "            plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                     arr[arr['degree'] == i]['basic_monthly_median'],\n",
    "                     color = 'blue', label = 'B')\n",
    "    plt.ylabel('basic_monthly_median')\n",
    "    plt.xlabel('year')\n",
    "    plt.xticks(list(arr['year']))\n",
    "    plt.title(str(uni) + ' basic_monthly_median salary between 2013 - 2017') # The for loop below ensures that the appropriate title gets called\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a generic function that can automatically generate mean salary plot for each university\n",
    "# A_Hons and B_Hons represent Accountancy and Business with Honours respectively\n",
    "# A_Cum Laude and B_Cum Laude represent Accountancy and Business with Cum Laude and above respectively (Exclusive to SMU)\n",
    "# A and B could mean Accountancy and Business without Hons/ Cum Laude depending on the university\n",
    "\n",
    "def gen_mean_salary(arr, unq_deg):\n",
    "    for i in unq_deg:\n",
    "        if '(Hons)' in i.split(): # NTU and NUS have (Hons) cases\n",
    "            if '(Accountancy)' in i.split() or 'Accountancy' in i.split(): # NUS has bracket while NTU does not\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['basic_monthly_mean'],\n",
    "                         color = 'red', linestyle = '--', label = 'A_Hons')               \n",
    "            elif 'Business' in i.split():\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['basic_monthly_mean'],\n",
    "                         color = 'blue', linestyle = '--', label = 'B_Hons')                \n",
    "        elif 'Laude' in i.split(): # For SMU Cum Laude \n",
    "            if 'Accountancy' in i.split():\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['basic_monthly_mean'],\n",
    "                         color = 'red', linestyle = '--', label = 'A_Cum Laude')\n",
    "            elif 'Business' in i.split():\n",
    "                plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                         arr[arr['degree'] == i]['basic_monthly_mean'],\n",
    "                         color = 'blue', linestyle = '--', label = 'B_Cum Laude')\n",
    "        elif '(Accountancy)' in i.split() or 'Accountancy' in i.split(): # NUS Accountancy has bracket\n",
    "            plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                     arr[arr['degree'] == i]['basic_monthly_mean'],\n",
    "                     color = 'red', label = 'A')\n",
    "        elif 'Business' in i.split(): # For NUS Business w/o (Hons) and SMU Business w/o Cum Laude \n",
    "            plt.plot(arr[arr['degree'] == i]['year'],\n",
    "                     arr[arr['degree'] == i]['basic_monthly_mean'],\n",
    "                     color = 'blue', label = 'B')\n",
    "    plt.ylabel('basic_monthly_mean')\n",
    "    plt.xlabel('year')\n",
    "    plt.xticks(list(arr['year']))\n",
    "    plt.title(str(uni) + ' basic_monthly_mean salary between 2013 - 2017') # The for loop in the next cell ensures that the appropriate title gets called\n",
    "    plt.legend()\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    },
    {
     "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": [
    "# Use a for loop to call all 3 plots for each university \n",
    "\n",
    "for uni in ['NTU', 'NUS', 'SMU']:\n",
    "    arr = df_comb[df_comb['university'] == uni]\n",
    "    unq_deg = list(arr['degree'].unique())\n",
    "    gen_emp_rate(arr, unq_deg)\n",
    "    gen_median_salary(arr, unq_deg)\n",
    "    gen_mean_salary(arr, unq_deg)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Insights\n",
    "\n",
    "While the analysis is fairly simple, it has revealed some very interesting insights based on how the trends have moved for the past 5 years excluding 2018 which we do not have.\n",
    "\n",
    "Insight 1) The employment rate for Accountancy graduates in each university have been consistently better than Business graduates. Could there be companies mass hiring Accountancy graduates upon their graduation? Perhaps, the big four (E&Y, Deloitte, PwC, KPMG)? Not to forget that majority of the Accountancy students would at least have interned at one of the big fours upon graduation\n",
    "\n",
    "Insight 2) While more Accountancy graduates are getting employed, their starting salaries have been consistently lower than Business graduates. For NUS, it even reached a point where the median salary of Business graduates without Hons were the same as Accountancy graduates with Hons in 2015 and 2016\n",
    "\n",
    "Insight 3) For the first time in 5 years, the employment rate for NTU Accountancy graduates have declined. Meanwhile, the employment rate for SMU Accountancy have shown signs of decline and slowing down. NUS however, is the only one where the emplyoment rate for its Accountancy graduates have been improving.\n",
    "\n",
    "Insight 4) Out of the 3 universities and in terms of median salary, NTU Business graduates have been getting paid the least where the all time high was 3300 in 2017 as compared to 3400 and 3800 for SMU (w/o cum laude, 2017) and NUS (hons, 2015) respectively. SMU w/o cum laude include the entire cohort ie cum laude and below. However, we must not forget that SMU include double degree students as well.\n",
    "\n",
    "Insight 5) The median salaries of Accountancy graduates regardless of universities have been very close (NTU_A_hons vs NUS_A_hons vs SMU_A). This could suggest that an average Accountancy graduates would find it hard to stand out. Accountancy graduates from SMU with Cum Laude were much more valued.\n",
    "\n",
    "Insight 6) Fortunately for both Accountancy and Business graduates, their median salaries are expected to steadily increase in the near future"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {
    "height": "638px",
    "width": "864px"
   },
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": true,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": true,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}