LivingIncomeAnalysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Living Income Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Summary:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook analyze whether cocoa farmers in Ghana and Côte d'Ivoire reach a Living Income.\n",
    "\n",
    "I use as starting point the [dataset](https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/82TWZJ) from the [Demystifying the Cocoa study](https://www.kit.nl/project/demystifying-cocoa-sector/), and build on the work published [here](https://c69aa8ac-6965-42b2-abb7-0f0b86c23d2e.filesusr.com/ugd/0c5ab3_fc3386a550b94a898c7757ee13ab59e6.pdf) and [here](https://c69aa8ac-6965-42b2-abb7-0f0b86c23d2e.filesusr.com/ugd/0c5ab3_93560a9b816d40c3a28daaa686e972a5.pdf)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this notebook, specifically I will:\n",
    "* Load the data\n",
    "* Do an initial variable and feature selection from the original dataset\n",
    "* Create Analytical groups\n",
    "* Calculate household income, performing data imputation where needed\n",
    "* Create a variable with the Living Income value per household and an indicator variable on whether this is achieved\n",
    "* Analyze the key indicators\n",
    "* Clean the remainder of the dataset\n",
    "* Prepare the dataset for Machine Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 0. Initial (Data) processing "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 0.1 Preamble"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First I load all libraries to be used:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "# magic word for producing visualizations in notebook\n",
    "%matplotlib inline\n",
    "\n",
    "import seaborn as sns#; sns.set(style=\"ticks\", color_codes=True)\n",
    "import json"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 0.2 Data Import and initial feature  & observation selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Load the data from stata, which has already been partially cleaned/treated, since it comes from a curated public data repository.\n",
    "\n",
    "The dataset is extracted from https://dataverse.harvard.edu/dataset.xhtml?persistentId=doi:10.7910/DVN/82TWZJ\n",
    "in stata 14 binary format.\n",
    "\n",
    "It is part of the research project described more broadly at https://www.kit.nl/project/demystifying-cocoa-sector/"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = pd.read_stata(\"Cocoa_Livelihoods_export_Dataverse.dta\") # data\n",
    "data_properties = pd.read_stata(\"Cocoa_Livelihoods_export_Dataverse.dta\", iterator = True) #meta data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I initialliy inspect the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of data set: (3045, 1252)\n"
     ]
    }
   ],
   "source": [
    "print(\"Shape of data set:\", data.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First few rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>today</th>\n",
       "      <th>country</th>\n",
       "      <th>region_comparable</th>\n",
       "      <th>up_region</th>\n",
       "      <th>region</th>\n",
       "      <th>sub_region</th>\n",
       "      <th>village</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>consent_yn</th>\n",
       "      <th>...</th>\n",
       "      <th>foodsec_bin_3</th>\n",
       "      <th>foodsec_bin_4</th>\n",
       "      <th>foodsec_bin_5</th>\n",
       "      <th>foodsec_bin_6</th>\n",
       "      <th>foodsec_bin_7</th>\n",
       "      <th>foodsec_bin_8</th>\n",
       "      <th>foodsec_bin_9</th>\n",
       "      <th>foodsec_bin_10</th>\n",
       "      <th>foodsec_bin_11</th>\n",
       "      <th>foodsec_bin_12</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>26-dec-16</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Atwima Mponua</td>\n",
       "      <td>Agogoso</td>\n",
       "      <td>6.61</td>\n",
       "      <td>-2.07</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>27-dec-16</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Atwima Nwabiagya</td>\n",
       "      <td>Toase</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>04-dec-16</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Adansi West</td>\n",
       "      <td>Ampunyasi</td>\n",
       "      <td>6.04</td>\n",
       "      <td>-1.76</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>03-dec-16</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Bosomtwi Kwanwoma</td>\n",
       "      <td>Nyameani</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>03-dec-16</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Bosomtwi Kwanwoma</td>\n",
       "      <td>Nyameani</td>\n",
       "      <td>6.57</td>\n",
       "      <td>-1.43</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>2-dec-2016</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Adansi East</td>\n",
       "      <td>Aboabo 2</td>\n",
       "      <td>6.22</td>\n",
       "      <td>-1.35</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>03-dec-16</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Bosomtwi Kwanwoma</td>\n",
       "      <td>Nyameani</td>\n",
       "      <td>6.57</td>\n",
       "      <td>-1.43</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>27-dec-16</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Atwima Nwabiagya</td>\n",
       "      <td>Toase</td>\n",
       "      <td>6.66</td>\n",
       "      <td>-1.81</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>29-dec-16</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Afigya Sekyere</td>\n",
       "      <td>Jamaase</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>04-dec-16</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Ghana</td>\n",
       "      <td>Ashanti</td>\n",
       "      <td>Adansi West</td>\n",
       "      <td>Ampunyasi</td>\n",
       "      <td>6.04</td>\n",
       "      <td>-1.76</td>\n",
       "      <td>yes</td>\n",
       "      <td>...</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "      <td>no</td>\n",
       "      <td>yes</td>\n",
       "      <td>yes</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>10 rows × 1252 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        today country region_comparable up_region   region         sub_region  \\\n",
       "0   26-dec-16   Ghana           Ashanti     Ghana  Ashanti      Atwima Mponua   \n",
       "1   27-dec-16   Ghana           Ashanti     Ghana  Ashanti   Atwima Nwabiagya   \n",
       "2   04-dec-16   Ghana           Ashanti     Ghana  Ashanti        Adansi West   \n",
       "3   03-dec-16   Ghana           Ashanti     Ghana  Ashanti  Bosomtwi Kwanwoma   \n",
       "4   03-dec-16   Ghana           Ashanti     Ghana  Ashanti  Bosomtwi Kwanwoma   \n",
       "5  2-dec-2016   Ghana           Ashanti     Ghana  Ashanti        Adansi East   \n",
       "6   03-dec-16   Ghana           Ashanti     Ghana  Ashanti  Bosomtwi Kwanwoma   \n",
       "7   27-dec-16   Ghana           Ashanti     Ghana  Ashanti   Atwima Nwabiagya   \n",
       "8   29-dec-16   Ghana           Ashanti     Ghana  Ashanti     Afigya Sekyere   \n",
       "9   04-dec-16   Ghana           Ashanti     Ghana  Ashanti        Adansi West   \n",
       "\n",
       "     village  latitude  longitude consent_yn  ...  foodsec_bin_3  \\\n",
       "0    Agogoso      6.61      -2.07        yes  ...            yes   \n",
       "1      Toase       NaN        NaN        yes  ...            yes   \n",
       "2  Ampunyasi      6.04      -1.76        yes  ...            yes   \n",
       "3   Nyameani       NaN        NaN        yes  ...            yes   \n",
       "4   Nyameani      6.57      -1.43        yes  ...            yes   \n",
       "5   Aboabo 2      6.22      -1.35        yes  ...             no   \n",
       "6   Nyameani      6.57      -1.43        yes  ...             no   \n",
       "7      Toase      6.66      -1.81        yes  ...            yes   \n",
       "8    Jamaase       NaN        NaN        yes  ...            yes   \n",
       "9  Ampunyasi      6.04      -1.76        yes  ...             no   \n",
       "\n",
       "   foodsec_bin_4  foodsec_bin_5  foodsec_bin_6  foodsec_bin_7  foodsec_bin_8  \\\n",
       "0            yes            yes            yes            yes            yes   \n",
       "1            yes            yes            yes            yes            yes   \n",
       "2            yes            yes            yes            yes            yes   \n",
       "3            yes            yes            yes            yes            yes   \n",
       "4            yes            yes            yes            yes            yes   \n",
       "5             no            yes            yes            yes            yes   \n",
       "6             no             no             no             no             no   \n",
       "7            yes            yes            yes            yes            yes   \n",
       "8            yes            yes            yes            yes            yes   \n",
       "9             no             no             no            yes            yes   \n",
       "\n",
       "   foodsec_bin_9  foodsec_bin_10  foodsec_bin_11  foodsec_bin_12  \n",
       "0            yes             yes             yes             yes  \n",
       "1            yes             yes             yes             yes  \n",
       "2            yes             yes             yes             yes  \n",
       "3            yes             yes             yes             yes  \n",
       "4            yes             yes             yes             yes  \n",
       "5            yes             yes             yes             yes  \n",
       "6             no              no              no              no  \n",
       "7            yes             yes             yes             yes  \n",
       "8            yes             yes             yes             yes  \n",
       "9            yes              no             yes             yes  \n",
       "\n",
       "[10 rows x 1252 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head(10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "See some descriptive statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>hhmem_number</th>\n",
       "      <th>male_adults_over_65</th>\n",
       "      <th>female_adults_over_65</th>\n",
       "      <th>male_adults_15_65</th>\n",
       "      <th>female_adults_15_65</th>\n",
       "      <th>male_children_5_14</th>\n",
       "      <th>female_children_5_14</th>\n",
       "      <th>male_children_0_4</th>\n",
       "      <th>...</th>\n",
       "      <th>dhs_combined</th>\n",
       "      <th>PPI_SCORE</th>\n",
       "      <th>PPI_100_Nat</th>\n",
       "      <th>PPI_125dollar_day</th>\n",
       "      <th>PPI_200dollar_day</th>\n",
       "      <th>PPI_250dollar_day</th>\n",
       "      <th>PPI_190dollar_day</th>\n",
       "      <th>PPI_310dollar_day</th>\n",
       "      <th>non_cocoa_land</th>\n",
       "      <th>share_cocoa_land</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>2210.000000</td>\n",
       "      <td>2210.000000</td>\n",
       "      <td>3004.000000</td>\n",
       "      <td>3004.000000</td>\n",
       "      <td>3004.000000</td>\n",
       "      <td>3004.000000</td>\n",
       "      <td>3004.000000</td>\n",
       "      <td>3004.000000</td>\n",
       "      <td>3004.000000</td>\n",
       "      <td>3004.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2435.000000</td>\n",
       "      <td>2992.000000</td>\n",
       "      <td>2992.000000</td>\n",
       "      <td>2992.000000</td>\n",
       "      <td>2992.000000</td>\n",
       "      <td>2992.000000</td>\n",
       "      <td>1545.000000</td>\n",
       "      <td>1545.000000</td>\n",
       "      <td>1887.000000</td>\n",
       "      <td>1887.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>6.429661</td>\n",
       "      <td>-4.112733</td>\n",
       "      <td>6.264314</td>\n",
       "      <td>0.142144</td>\n",
       "      <td>0.147803</td>\n",
       "      <td>1.689747</td>\n",
       "      <td>1.705726</td>\n",
       "      <td>0.981025</td>\n",
       "      <td>0.868842</td>\n",
       "      <td>0.374501</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.291392</td>\n",
       "      <td>44.137032</td>\n",
       "      <td>30.232420</td>\n",
       "      <td>14.947255</td>\n",
       "      <td>33.256577</td>\n",
       "      <td>44.436817</td>\n",
       "      <td>7.543701</td>\n",
       "      <td>24.523375</td>\n",
       "      <td>1.705683</td>\n",
       "      <td>0.732727</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.697673</td>\n",
       "      <td>1.894443</td>\n",
       "      <td>3.200178</td>\n",
       "      <td>0.374114</td>\n",
       "      <td>0.402447</td>\n",
       "      <td>1.247421</td>\n",
       "      <td>1.181760</td>\n",
       "      <td>1.123111</td>\n",
       "      <td>1.099499</td>\n",
       "      <td>0.681247</td>\n",
       "      <td>...</td>\n",
       "      <td>0.487034</td>\n",
       "      <td>13.002094</td>\n",
       "      <td>22.378967</td>\n",
       "      <td>16.871199</td>\n",
       "      <td>28.074274</td>\n",
       "      <td>31.169250</td>\n",
       "      <td>9.881251</td>\n",
       "      <td>20.898497</td>\n",
       "      <td>2.309870</td>\n",
       "      <td>0.238368</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>4.840000</td>\n",
       "      <td>-8.200000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>-1.672072</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.052632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.940000</td>\n",
       "      <td>-5.447500</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.605459</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>11.700000</td>\n",
       "      <td>0.600000</td>\n",
       "      <td>5.300000</td>\n",
       "      <td>12.800000</td>\n",
       "      <td>0.900000</td>\n",
       "      <td>5.800000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.555556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>6.470000</td>\n",
       "      <td>-3.870000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.363614</td>\n",
       "      <td>43.000000</td>\n",
       "      <td>28.900000</td>\n",
       "      <td>7.200000</td>\n",
       "      <td>27.100000</td>\n",
       "      <td>41.299999</td>\n",
       "      <td>2.400000</td>\n",
       "      <td>16.100000</td>\n",
       "      <td>0.980000</td>\n",
       "      <td>0.750000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>6.930000</td>\n",
       "      <td>-2.500000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.055659</td>\n",
       "      <td>53.000000</td>\n",
       "      <td>50.700001</td>\n",
       "      <td>27.100000</td>\n",
       "      <td>59.400002</td>\n",
       "      <td>74.099998</td>\n",
       "      <td>7.900000</td>\n",
       "      <td>35.400002</td>\n",
       "      <td>2.296600</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>7.840000</td>\n",
       "      <td>-0.470000</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>2.926933</td>\n",
       "      <td>98.000000</td>\n",
       "      <td>87.599998</td>\n",
       "      <td>71.500000</td>\n",
       "      <td>93.900002</td>\n",
       "      <td>97.199997</td>\n",
       "      <td>50.700001</td>\n",
       "      <td>81.699997</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 347 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "          latitude    longitude  hhmem_number  male_adults_over_65  \\\n",
       "count  2210.000000  2210.000000   3004.000000          3004.000000   \n",
       "mean      6.429661    -4.112733      6.264314             0.142144   \n",
       "std       0.697673     1.894443      3.200178             0.374114   \n",
       "min       4.840000    -8.200000      1.000000             0.000000   \n",
       "25%       5.940000    -5.447500      4.000000             0.000000   \n",
       "50%       6.470000    -3.870000      6.000000             0.000000   \n",
       "75%       6.930000    -2.500000      8.000000             0.000000   \n",
       "max       7.840000    -0.470000     21.000000             5.000000   \n",
       "\n",
       "       female_adults_over_65  male_adults_15_65  female_adults_15_65  \\\n",
       "count            3004.000000        3004.000000          3004.000000   \n",
       "mean                0.147803           1.689747             1.705726   \n",
       "std                 0.402447           1.247421             1.181760   \n",
       "min                 0.000000           0.000000             0.000000   \n",
       "25%                 0.000000           1.000000             1.000000   \n",
       "50%                 0.000000           1.000000             1.000000   \n",
       "75%                 0.000000           2.000000             2.000000   \n",
       "max                 5.000000           9.000000             9.000000   \n",
       "\n",
       "       male_children_5_14  female_children_5_14  male_children_0_4  ...  \\\n",
       "count         3004.000000           3004.000000        3004.000000  ...   \n",
       "mean             0.981025              0.868842           0.374501  ...   \n",
       "std              1.123111              1.099499           0.681247  ...   \n",
       "min              0.000000              0.000000           0.000000  ...   \n",
       "25%              0.000000              0.000000           0.000000  ...   \n",
       "50%              1.000000              1.000000           0.000000  ...   \n",
       "75%              2.000000              1.000000           1.000000  ...   \n",
       "max              7.000000              7.000000           5.000000  ...   \n",
       "\n",
       "       dhs_combined    PPI_SCORE  PPI_100_Nat  PPI_125dollar_day  \\\n",
       "count   2435.000000  2992.000000  2992.000000        2992.000000   \n",
       "mean      -0.291392    44.137032    30.232420          14.947255   \n",
       "std        0.487034    13.002094    22.378967          16.871199   \n",
       "min       -1.672072    10.000000     0.000000           0.000000   \n",
       "25%       -0.605459    35.000000    11.700000           0.600000   \n",
       "50%       -0.363614    43.000000    28.900000           7.200000   \n",
       "75%       -0.055659    53.000000    50.700001          27.100000   \n",
       "max        2.926933    98.000000    87.599998          71.500000   \n",
       "\n",
       "       PPI_200dollar_day  PPI_250dollar_day  PPI_190dollar_day  \\\n",
       "count        2992.000000        2992.000000        1545.000000   \n",
       "mean           33.256577          44.436817           7.543701   \n",
       "std            28.074274          31.169250           9.881251   \n",
       "min             0.000000           0.000000           0.000000   \n",
       "25%             5.300000          12.800000           0.900000   \n",
       "50%            27.100000          41.299999           2.400000   \n",
       "75%            59.400002          74.099998           7.900000   \n",
       "max            93.900002          97.199997          50.700001   \n",
       "\n",
       "       PPI_310dollar_day  non_cocoa_land  share_cocoa_land  \n",
       "count        1545.000000     1887.000000       1887.000000  \n",
       "mean           24.523375        1.705683          0.732727  \n",
       "std            20.898497        2.309870          0.238368  \n",
       "min             0.000000        0.000000          0.052632  \n",
       "25%             5.800000        0.000000          0.555556  \n",
       "50%            16.100000        0.980000          0.750000  \n",
       "75%            35.400002        2.296600          1.000000  \n",
       "max            81.699997       17.000000          1.000000  \n",
       "\n",
       "[8 rows x 347 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Take a look at the full list of variables and variable labels:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{\n",
      " \"today\": \"Interview date\",\n",
      " \"country\": \"Country\",\n",
      " \"region_comparable\": \"Comparable regions\",\n",
      " \"up_region\": \"Country (Ghana) / District (CdI)\",\n",
      " \"region\": \"Region\",\n",
      " \"sub_region\": \"District (Ghana) / Departement (Cdi)\",\n",
      " \"village\": \"Village\",\n",
      " \"latitude\": \"Latitude\",\n",
      " \"longitude\": \"Longitude\",\n",
      " \"consent_yn\": \"Consent to participate\",\n",
      " \"hhmem_number\": \"Number of household members living in the compound\",\n",
      " \"male_adults_over_65\": \"Number of males over 65 years old\",\n",
      " \"female_adults_over_65\": \"Number of females over 65 years old\",\n",
      " \"male_adults_15_65\": \"Number of males 15-65\",\n",
      " \"female_adults_15_65\": \"Number of females between 15 and 65 years old\",\n",
      " \"male_children_5_14\": \"Number of boys between 15 and 65 years old\",\n",
      " \"female_children_5_14\": \"Number of girls  between 5 and 14 years old\",\n",
      " \"male_children_0_4\": \"Number of boys younger than 4 years old\",\n",
      " \"female_children_0_4\": \"Number of girls younger than 4 years old\",\n",
      " \"dep_ratio\": \"Dependency ratio\",\n",
      " \"head_age\": \"Head: age\",\n",
      " \"head_age_cat\": \"Head: age group\",\n",
      " \"head_gender\": \"Head: sex\",\n",
      " \"head_education\": \"Head: education level\",\n",
      " \"p1_respondent_age\": \"Respondent: age\",\n",
      " \"age_cat\": \"Respondent: age group\",\n",
      " \"p1_respondent_gender\": \"Respondent: sex\",\n",
      " \"p1_respondent_head_yn\": \"Respondent: head of the household\",\n",
      " \"p1_respondent_education\": \"Respondent: education level\",\n",
      " \"p1_respondent_read\": \"Respondent: hability to read in english/french\",\n",
      " \"p1_respondent_marital\": \"Respondent: marital status\",\n",
      " \"leader\": \"Respondent has a leadership position\",\n",
      " \"p1_leadership_role_0\": \"Leadership position: no / none\",\n",
      " \"p1_leadership_role_1\": \"Leadership position: chief farmer\",\n",
      " \"p1_leadership_role_2\": \"Leadership position: lead farmer\",\n",
      " \"p1_leadership_role_3\": \"Leadership position: purchasing clerk\",\n",
      " \"p1_leadership_role_4\": \"Leadership position: queen mother\",\n",
      " \"p1_leadership_role_5\": \"Leadership position: farmer organisation leader\",\n",
      " \"p1_leadership_role_6\": \"Leadership position: other association leader\",\n",
      " \"p1_leadership_role_7\": \"Leadership position: church / mosque leader\",\n",
      " \"p1_leadership_role_8\": \"Leadership position: extension officer / trainer\",\n",
      " \"p1_leadership_role_9\": \"Leadership position: womens leader\",\n",
      " \"p1_leadership_role_10\": \"Leadership position: opinion leader\",\n",
      " \"p1_leadership_role_11\": \"Leadership position: assembly man/women\",\n",
      " \"p1_leadership_role_12\": \"Leadership position: elder\",\n",
      " \"p1_leadership_role_13\": \"Leadership position: village chief\",\n",
      " \"p1_leadership_role_15\": \"Leadership position: tribe chief\",\n",
      " \"p1_leadership_role_16\": \"Leadership position: canton chief\",\n",
      " \"p1_leadership_role_17\": \"Leadership position: youth president\",\n",
      " \"p1_leadership_role_14\": \"Leadership position: other\",\n",
      " \"p1_respondent_born_district\": \"Respondent:  born in the same district\",\n",
      " \"p1_respondent_born_region\": \"Respondent:  born in the same region\",\n",
      " \"p1_respondent_born_country\": \"Respondent:  born in the same country\",\n",
      " \"p1_business_trade\": \"Respondent:  small business or trade\",\n",
      " \"p1_respondent_fem_firstchild\": \"Respondent:  age when first child was born\",\n",
      " \"male_children_5_14_school\": \"All household boys ages 5 to 14 currently attending school\",\n",
      " \"female_children_5_14_school\": \"All household girls ages 5 to 14 currently attending school\",\n",
      " \"incsource_biggest1\": \"Household's biggest source of income in last year\",\n",
      " \"incsource_biggest2\": \"Household's second biggest source of income in last year\",\n",
      " \"hh_income_perc_cocoa\": \"Estimated income percentage from sale of cocoa\",\n",
      " \"hh_income_perc_othercrops\": \"Estimated income percentage from sale of other crops\",\n",
      " \"hh_income_perc_livestock\": \"Estimated income percentage from sale of livestock or livestock products\",\n",
      " \"hh_income_perc_fish\": \"Estimated income percentage from sale of fish\",\n",
      " \"hh_income_perc_bushproducts\": \"Estimated income percentage from sale of bush products (bush meat, charcoal, woo\",\n",
      " \"hh_income_perc_farmlabour\": \"Estimated income percentage from labouring for other people on their farms\",\n",
      " \"hh_income_perc_nonaglabour\": \"Estimated income percentage from ;abouring for other people non-agriculture\",\n",
      " \"hh_income_perc_smallbusiness\": \"Estimated income percentage from own small business or trading\",\n",
      " \"hh_income_perc_comp\": \"Estimated income percentage from salary employment with a company\",\n",
      " \"hh_income_perc_govt\": \"Estimated income percentage from salary employment in government job (teacher, n\",\n",
      " \"hh_income_perc_landsalelease\": \"Estimated income percentage from sale or lease of land\",\n",
      " \"hh_income_perc_remittances\": \"Estimated income percentage from remittances from friends and family living away\",\n",
      " \"hh_income_perc_other\": \"Estimated income percentage from other\",\n",
      " \"crops_all_produced_1\": \"Produced avocado last season\",\n",
      " \"crops_all_produced_2\": \"Produced bananas last season\",\n",
      " \"crops_all_produced_3\": \"Produced soya beans last season\",\n",
      " \"crops_all_produced_4\": \"Produced beans last season\",\n",
      " \"crops_all_produced_5\": \"Produced cabbages last season\",\n",
      " \"crops_all_produced_6\": \"Produced carrots last season\",\n",
      " \"crops_all_produced_7\": \"Produced cashews last season\",\n",
      " \"crops_all_produced_8\": \"Produced cassava last season\",\n",
      " \"crops_all_produced_9\": \"Produced chili last season\",\n",
      " \"crops_all_produced_10\": \"Produced cocoa last season\",\n",
      " \"crops_all_produced_11\": \"Produced coconut last season\",\n",
      " \"crops_all_produced_12\": \"Produced cocoyam last season\",\n",
      " \"crops_all_produced_13\": \"Produced coffee last season\",\n",
      " \"crops_all_produced_14\": \"Produced courgette last season\",\n",
      " \"crops_all_produced_15\": \"Produced cotton last season\",\n",
      " \"crops_all_produced_16\": \"Produced cow pea last season\",\n",
      " \"crops_all_produced_17\": \"Produced eggplant last season\",\n",
      " \"crops_all_produced_18\": \"Produced fodder last season\",\n",
      " \"crops_all_produced_19\": \"Produced garlic last season\",\n",
      " \"crops_all_produced_20\": \"Produced ginger last season\",\n",
      " \"crops_all_produced_21\": \"Produced groundnuts last season\",\n",
      " \"crops_all_produced_22\": \"Produced lemon last season\",\n",
      " \"crops_all_produced_23\": \"Produced lime last season\",\n",
      " \"crops_all_produced_24\": \"Produced maize last season\",\n",
      " \"crops_all_produced_25\": \"Produced mango last season\",\n",
      " \"crops_all_produced_26\": \"Produced millet last season\",\n",
      " \"crops_all_produced_27\": \"Produced okra last season\",\n",
      " \"crops_all_produced_28\": \"Produced onions last season\",\n",
      " \"crops_all_produced_29\": \"Produced oranges last season\",\n",
      " \"crops_all_produced_30\": \"Produced palm last season\",\n",
      " \"crops_all_produced_31\": \"Produced papaya last season\",\n",
      " \"crops_all_produced_32\": \"Produced peas last season\",\n",
      " \"crops_all_produced_33\": \"Produced peppers last season\",\n",
      " \"crops_all_produced_34\": \"Produced pineapple last season\",\n",
      " \"crops_all_produced_35\": \"Produced plantain last season\",\n",
      " \"crops_all_produced_36\": \"Produced irish potato last season\",\n",
      " \"crops_all_produced_37\": \"Produced sweet potato last season\",\n",
      " \"crops_all_produced_38\": \"Produced rice last season\",\n",
      " \"crops_all_produced_39\": \"Produced rubber last season\",\n",
      " \"crops_all_produced_40\": \"Produced spinach last season\",\n",
      " \"crops_all_produced_41\": \"Produced sorghum last season\",\n",
      " \"crops_all_produced_42\": \"Produced shea nut last season\",\n",
      " \"crops_all_produced_43\": \"Produced sugar cane last season\",\n",
      " \"crops_all_produced_44\": \"Produced tangerine last season\",\n",
      " \"crops_all_produced_45\": \"Produced tomatoes last season\",\n",
      " \"crops_all_produced_46\": \"Produced watermelon last season\",\n",
      " \"crops_all_produced_47\": \"Produced yam last season\",\n",
      " \"crops_all_produced_48\": \"Produced other last season\",\n",
      " \"crops_all_produced_49\": \"Produced none/ not applicable last season\",\n",
      " \"n_crops_produced\": \"Number of crops produced\",\n",
      " \"crops_all_sold_1\": \"Sold avocado last season\",\n",
      " \"crops_all_sold_2\": \"Sold bananas last season\",\n",
      " \"crops_all_sold_3\": \"Sold soya beans last season\",\n",
      " \"crops_all_sold_4\": \"Sold beans last season\",\n",
      " \"crops_all_sold_5\": \"Sold cabbages last season\",\n",
      " \"crops_all_sold_6\": \"Sold carrots last season\",\n",
      " \"crops_all_sold_7\": \"Sold cashews last season\",\n",
      " \"crops_all_sold_8\": \"Sold cassava last season\",\n",
      " \"crops_all_sold_9\": \"Sold chili last season\",\n",
      " \"crops_all_sold_10\": \"Sold cocoa last season\",\n",
      " \"crops_all_sold_11\": \"Sold coconut last season\",\n",
      " \"crops_all_sold_12\": \"Sold cocoyam last season\",\n",
      " \"crops_all_sold_13\": \"Sold coffee last season\",\n",
      " \"crops_all_sold_14\": \"Sold courgette last season\",\n",
      " \"crops_all_sold_15\": \"Sold cotton last season\",\n",
      " \"crops_all_sold_16\": \"Sold cow pea last season\",\n",
      " \"crops_all_sold_17\": \"Sold eggplant last season\",\n",
      " \"crops_all_sold_18\": \"Sold fodder last season\",\n",
      " \"crops_all_sold_19\": \"Sold garlic last season\",\n",
      " \"crops_all_sold_20\": \"Sold ginger last season\",\n",
      " \"crops_all_sold_21\": \"Sold groundnuts last season\",\n",
      " \"crops_all_sold_22\": \"Sold lemon last season\",\n",
      " \"crops_all_sold_23\": \"Sold lime last season\",\n",
      " \"crops_all_sold_24\": \"Sold maize last season\",\n",
      " \"crops_all_sold_25\": \"Sold mango last season\",\n",
      " \"crops_all_sold_26\": \"Sold millet last season\",\n",
      " \"crops_all_sold_27\": \"Sold okra last season\",\n",
      " \"crops_all_sold_28\": \"Sold onions last season\",\n",
      " \"crops_all_sold_29\": \"Sold oranges last season\",\n",
      " \"crops_all_sold_30\": \"Sold palm last season\",\n",
      " \"crops_all_sold_31\": \"Sold papaya last season\",\n",
      " \"crops_all_sold_32\": \"Sold peas last season\",\n",
      " \"crops_all_sold_33\": \"Sold peppers last season\",\n",
      " \"crops_all_sold_34\": \"Sold pineapple last season\",\n",
      " \"crops_all_sold_35\": \"Sold plantain last season\",\n",
      " \"crops_all_sold_36\": \"Sold irish potato last season\",\n",
      " \"crops_all_sold_37\": \"Sold sweet potato last season\",\n",
      " \"crops_all_sold_38\": \"Sold rice last season\",\n",
      " \"crops_all_sold_39\": \"Sold rubber last season\",\n",
      " \"crops_all_sold_40\": \"Sold spinach last season\",\n",
      " \"crops_all_sold_41\": \"Sold sorghum last season\",\n",
      " \"crops_all_sold_42\": \"Sold shea nut last season\",\n",
      " \"crops_all_sold_43\": \"Sold sugar cane last season\",\n",
      " \"crops_all_sold_44\": \"Sold tangerine last season\",\n",
      " \"crops_all_sold_45\": \"Sold tomatoes last season\",\n",
      " \"crops_all_sold_46\": \"Sold watermelon last season\",\n",
      " \"crops_all_sold_47\": \"Sold yam last season\",\n",
      " \"crops_all_sold_48\": \"Sold other last season\",\n",
      " \"crops_all_sold_49\": \"Sold none/ not applicable last season\",\n",
      " \"n_crops_sold\": \"Sold number of crops sold last season\",\n",
      " \"crops_sold_ifprod_1\": \"Sold avocado last season\",\n",
      " \"crops_sold_ifprod_2\": \"Sold bananas last season\",\n",
      " \"crops_sold_ifprod_3\": \"Sold soya beans last season\",\n",
      " \"crops_sold_ifprod_4\": \"Sold beans last season\",\n",
      " \"crops_sold_ifprod_5\": \"Sold cabbages last season\",\n",
      " \"crops_sold_ifprod_6\": \"Sold carrots last season\",\n",
      " \"crops_sold_ifprod_7\": \"Sold cashews last season\",\n",
      " \"crops_sold_ifprod_8\": \"Sold cassava last season\",\n",
      " \"crops_sold_ifprod_9\": \"Sold chili last season\",\n",
      " \"crops_sold_ifprod_10\": \"Sold cocoa last season\",\n",
      " \"crops_sold_ifprod_11\": \"Sold coconut last season\",\n",
      " \"crops_sold_ifprod_12\": \"Sold cocoyam last season\",\n",
      " \"crops_sold_ifprod_13\": \"Sold coffee last season\",\n",
      " \"crops_sold_ifprod_14\": \"Sold courgette last season\",\n",
      " \"crops_sold_ifprod_15\": \"Sold cotton last season\",\n",
      " \"crops_sold_ifprod_16\": \"Sold cow pea last season\",\n",
      " \"crops_sold_ifprod_17\": \"Sold eggplant last season\",\n",
      " \"crops_sold_ifprod_18\": \"Sold fodder last season\",\n",
      " \"crops_sold_ifprod_19\": \"Sold garlic last season\",\n",
      " \"crops_sold_ifprod_20\": \"Sold ginger last season\",\n",
      " \"crops_sold_ifprod_21\": \"Sold groundnuts last season\",\n",
      " \"crops_sold_ifprod_22\": \"Sold lemon last season\",\n",
      " \"crops_sold_ifprod_23\": \"Sold lime last season\",\n",
      " \"crops_sold_ifprod_24\": \"Sold maize last season\",\n",
      " \"crops_sold_ifprod_25\": \"Sold mango last season\",\n",
      " \"crops_sold_ifprod_26\": \"Sold millet last season\",\n",
      " \"crops_sold_ifprod_27\": \"Sold okra last season\",\n",
      " \"crops_sold_ifprod_28\": \"Sold onions last season\",\n",
      " \"crops_sold_ifprod_29\": \"Sold oranges last season\",\n",
      " \"crops_sold_ifprod_30\": \"Sold palm last season\",\n",
      " \"crops_sold_ifprod_31\": \"Sold papaya last season\",\n",
      " \"crops_sold_ifprod_32\": \"Sold peas last season\",\n",
      " \"crops_sold_ifprod_33\": \"Sold peppers last season\",\n",
      " \"crops_sold_ifprod_34\": \"Sold pineapple last season\",\n",
      " \"crops_sold_ifprod_35\": \"Sold plantain last season\",\n",
      " \"crops_sold_ifprod_36\": \"Sold irish potato last season\",\n",
      " \"crops_sold_ifprod_37\": \"Sold sweet potato last season\",\n",
      " \"crops_sold_ifprod_38\": \"Sold rice last season\",\n",
      " \"crops_sold_ifprod_39\": \"Sold rubber last season\",\n",
      " \"crops_sold_ifprod_40\": \"Sold spinach last season\",\n",
      " \"crops_sold_ifprod_41\": \"Sold sorghum last season\",\n",
      " \"crops_sold_ifprod_42\": \"Sold shea nut last season\",\n",
      " \"crops_sold_ifprod_43\": \"Sold sugar cane last season\",\n",
      " \"crops_sold_ifprod_44\": \"Sold tangerine last season\",\n",
      " \"crops_sold_ifprod_45\": \"Sold tomatoes last season\",\n",
      " \"crops_sold_ifprod_46\": \"Sold watermelon last season\",\n",
      " \"crops_sold_ifprod_47\": \"Sold yam last season\",\n",
      " \"crops_important1\": \"Most important crop\",\n",
      " \"crops_important2\": \"Second most important crop\",\n",
      " \"cocoa_hh\": \"Cocoa household\",\n",
      " \"coconut_hh\": \"Coconut household\",\n",
      " \"cocoyam_hh\": \"Cocoyam household\",\n",
      " \"coffee_hh\": \"Coffee household\",\n",
      " \"courgette_hh\": \"Courgette household\",\n",
      " \"cotton_hh\": \"Cotton household\",\n",
      " \"cow_pea_hh\": \"Cow pea household\",\n",
      " \"eggplant_hh\": \"Eggplant household\",\n",
      " \"fodder_hh\": \"Fodder household\",\n",
      " \"garlic_hh\": \"Garlic household\",\n",
      " \"ginger_hh\": \"Ginger household\",\n",
      " \"groundnuts_hh\": \"Groundnuts household\",\n",
      " \"lemon_hh\": \"Lemon household\",\n",
      " \"lime_hh\": \"Lime household\",\n",
      " \"maize_hh\": \"Maize household\",\n",
      " \"mango_hh\": \"Mango household\",\n",
      " \"millet_hh\": \"Millet household\",\n",
      " \"okra_hh\": \"Okra household\",\n",
      " \"onions_hh\": \"Onions household\",\n",
      " \"oranges_hh\": \"Oranges household\",\n",
      " \"palm_hh\": \"Palm household\",\n",
      " \"papaya_hh\": \"Papaya household\",\n",
      " \"peas_hh\": \"Peas household\",\n",
      " \"peppers_hh\": \"Peppers household\",\n",
      " \"pineapple_hh\": \"Pineapple household\",\n",
      " \"plantai_hh\": \"Plantain household\",\n",
      " \"irish_potato_hh\": \"Irish potato household\",\n",
      " \"sweet_potato_hh\": \"Sweet potato household\",\n",
      " \"rice_hh\": \"Rice household\",\n",
      " \"rubber_hh\": \"Rubber household\",\n",
      " \"spinach_hh\": \"Spinach household\",\n",
      " \"sorghum_hh\": \"Sorghum household\",\n",
      " \"shea_nut_hh\": \"Shea nut household\",\n",
      " \"sugar_cane_hh\": \"Sugar cane household\",\n",
      " \"tangerine_hh\": \"Tangerine household\",\n",
      " \"tomatoes_hh\": \"Tomatoes household\",\n",
      " \"watermelon_hh\": \"Watermelon household\",\n",
      " \"yam_hh\": \"Yam household\",\n",
      " \"crops_important1_5ago\": \"Most important crop, 5 years ago\",\n",
      " \"crops_important2_5ago\": \"Second most important crop, 5 years ago\",\n",
      " \"avocado_hh_5ago\": \"Avocado household, 5 years ago\",\n",
      " \"bananas_hh_5ago\": \"Bananas household, 5 years ago\",\n",
      " \"soya_beans_hh_5ago\": \"Soya beans household, 5 years ago\",\n",
      " \"beans_hh_5ago\": \"Beans household, 5 years ago\",\n",
      " \"cabbages_hh_5ago\": \"Cabbages household, 5 years ago\",\n",
      " \"carrots_hh_5ago\": \"Carrots household, 5 years ago\",\n",
      " \"cashews_hh_5ago\": \"Cashews household, 5 years ago\",\n",
      " \"cassava_hh_5ago\": \"Cassava household, 5 years ago\",\n",
      " \"chili_hh_5ago\": \"Chili household, 5 years ago\",\n",
      " \"cocoa_hh_5ago\": \"Cocoa household, 5 years ago\",\n",
      " \"coconut_hh_5ago\": \"Coconut household, 5 years ago\",\n",
      " \"cocoyam_hh_5ago\": \"Cocoyam household, 5 years ago\",\n",
      " \"coffee_hh_5ago\": \"Coffee household, 5 years ago\",\n",
      " \"courgette_hh_5ago\": \"Courgette household, 5 years ago\",\n",
      " \"cotton_hh_5ago\": \"Cotton household, 5 years ago\",\n",
      " \"cow_pea_hh_5ago\": \"Cow pea household, 5 years ago\",\n",
      " \"eggplant_hh_5ago\": \"Eggplant household, 5 years ago\",\n",
      " \"fodder_hh_5ago\": \"Fodder household, 5 years ago\",\n",
      " \"garlic_hh_5ago\": \"Garlic household, 5 years ago\",\n",
      " \"ginger_hh_5ago\": \"Ginger household, 5 years ago\",\n",
      " \"groundnuts_hh_5ago\": \"Groundnuts household, 5 years ago\",\n",
      " \"lemon_hh_5ago\": \"Lemon household, 5 years ago\",\n",
      " \"lime_hh_5ago\": \"Lime household, 5 years ago\",\n",
      " \"maize_hh_5ago\": \"Maize household, 5 years ago\",\n",
      " \"mango_hh_5ago\": \"Mango household, 5 years ago\",\n",
      " \"millet_hh_5ago\": \"Millet household, 5 years ago\",\n",
      " \"okra_hh_5ago\": \"Okra household, 5 years ago\",\n",
      " \"onions_hh_5ago\": \"Onions household, 5 years ago\",\n",
      " \"oranges_hh_5ago\": \"Oranges household, 5 years ago\",\n",
      " \"palm_hh_5ago\": \"Palm household, 5 years ago\",\n",
      " \"papaya_hh_5ago\": \"Papaya household, 5 years ago\",\n",
      " \"peas_hh_5ago\": \"Peas household, 5 years ago\",\n",
      " \"peppers_hh_5ago\": \"Peppers household, 5 years ago\",\n",
      " \"pineapple_hh_5ago\": \"Pineapple household, 5 years ago\",\n",
      " \"plantai_hh_5ago\": \"Plantain household, 5 years ago\",\n",
      " \"irish_potato_hh_5ago\": \"Irish potato household, 5 years ago\",\n",
      " \"sweet_potato_hh_5ago\": \"Sweet potato household, 5 years ago\",\n",
      " \"rice_hh_5ago\": \"Rice household, 5 years ago\",\n",
      " \"rubber_hh_5ago\": \"Rubber household, 5 years ago\",\n",
      " \"spinach_hh_5ago\": \"Spinach household, 5 years ago\",\n",
      " \"sorghum_hh_5ago\": \"Sorghum household, 5 years ago\",\n",
      " \"shea_nut_hh_5ago\": \"Shea nut household, 5 years ago\",\n",
      " \"sugar_cane_hh_5ago\": \"Sugar cane household, 5 years ago\",\n",
      " \"tangerine_hh_5ago\": \"Tangerine household, 5 years ago\",\n",
      " \"tomatoes_hh_5ago\": \"Tomatoes household, 5 years ago\",\n",
      " \"watermelon_hh_5ago\": \"Watermelon household, 5 years ago\",\n",
      " \"yam_hh_5ago\": \"Yam household, 5 years ago\",\n",
      " \"avocado_hh_change\": \"Moved in/out as avocado household\",\n",
      " \"bananas_hh_change\": \"Moved in/out as bananas household\",\n",
      " \"soya_beans_hh_change\": \"Moved in/out as soya beans household\",\n",
      " \"beans_hh_change\": \"Moved in/out as beans household\",\n",
      " \"cabbages_hh_change\": \"Moved in/out as cabbages household\",\n",
      " \"carrots_hh_change\": \"Moved in/out as carrots household\",\n",
      " \"cashews_hh_change\": \"Moved in/out as cashews household\",\n",
      " \"cassava_hh_change\": \"Moved in/out as cassava household\",\n",
      " \"chili_hh_change\": \"Moved in/out as chili household\",\n",
      " \"cocoa_hh_change\": \"Moved in/out as cocoa household\",\n",
      " \"coconut_hh_change\": \"Moved in/out as coconut household\",\n",
      " \"cocoyam_hh_change\": \"Moved in/out as cocoyam household\",\n",
      " \"coffee_hh_change\": \"Moved in/out as coffee household\",\n",
      " \"courgette_hh_change\": \"Moved in/out as courgette household\",\n",
      " \"cotton_hh_change\": \"Moved in/out as cotton household\",\n",
      " \"cow_pea_hh_change\": \"Moved in/out as cow pea household\",\n",
      " \"eggplant_hh_change\": \"Moved in/out as eggplant household\",\n",
      " \"fodder_hh_change\": \"Moved in/out as fodder household\",\n",
      " \"garlic_hh_change\": \"Moved in/out as garlic household\",\n",
      " \"ginger_hh_change\": \"Moved in/out as ginger household\",\n",
      " \"groundnuts_hh_change\": \"Moved in/out as groundnuts household\",\n",
      " \"lemon_hh_change\": \"Moved in/out as lemon household\",\n",
      " \"lime_hh_change\": \"Moved in/out as lime household\",\n",
      " \"maize_hh_change\": \"Moved in/out as maize household\",\n",
      " \"mango_hh_change\": \"Moved in/out as mango household\",\n",
      " \"millet_hh_change\": \"Moved in/out as millet household\",\n",
      " \"okra_hh_change\": \"Moved in/out as okra household\",\n",
      " \"onions_hh_change\": \"Moved in/out as onions household\",\n",
      " \"oranges_hh_change\": \"Moved in/out as oranges household\",\n",
      " \"palm_hh_change\": \"Moved in/out as palm household\",\n",
      " \"papaya_hh_change\": \"Moved in/out as papaya household\",\n",
      " \"peas_hh_change\": \"Moved in/out as peas household\",\n",
      " \"peppers_hh_change\": \"Moved in/out as peppers household\",\n",
      " \"pineapple_hh_change\": \"Moved in/out as pineapple household\",\n",
      " \"plantai_hh_change\": \"Moved in/out as plantain household\",\n",
      " \"irish_potato_hh_change\": \"Moved in/out as irish potato household\",\n",
      " \"sweet_potato_hh_change\": \"Moved in/out as sweet potato household\",\n",
      " \"rice_hh_change\": \"Moved in/out as rice household\",\n",
      " \"rubber_hh_change\": \"Moved in/out as rubber household\",\n",
      " \"spinach_hh_change\": \"Moved in/out as spinach household\",\n",
      " \"sorghum_hh_change\": \"Moved in/out as sorghum household\",\n",
      " \"shea_nut_hh_change\": \"Moved in/out as shea nut household\",\n",
      " \"sugar_cane_hh_change\": \"Moved in/out as sugar cane household\",\n",
      " \"tangerine_hh_change\": \"Moved in/out as tangerine household\",\n",
      " \"tomatoes_hh_change\": \"Moved in/out as tomatoes household\",\n",
      " \"watermelon_hh_change\": \"Moved in/out as watermelon household\",\n",
      " \"yam_hh_change\": \"Moved in/out as yam household\",\n",
      " \"land_unit_preference\": \"Unit for land\",\n",
      " \"all_land_known_yn\": \"Do you know how much land was used to cultivate all crops last year?\",\n",
      " \"all_land_cultivated_ha\": \"Land used to cultivate all crops (ha)\",\n",
      " \"all_land_owned_ha\": \"All land owned (ha)\",\n",
      " \"all_land_leased_ha\": \"All land leased (ha)\",\n",
      " \"all_land_scrop_abunu_ha\": \"All land under abunu sharecropping arrangement (ha)\",\n",
      " \"all_land_scrop_abusa_ha\": \"All land under abusa sharecropping arrangement (ha)\",\n",
      " \"all_land_fallow_ha\": \"All fallow land or land left for grazing (ha)\",\n",
      " \"all_land_numlandparcels\": \"Total number of plots\",\n",
      " \"all_land_leasedout_yn\": \"Land leased out\",\n",
      " \"all_land_leasedout_landunit_ha\": \"Land leased out (ha)\",\n",
      " \"cocoa_land_known_yn\": \"Do you know how much land was used to cultivate cocoa last year?\",\n",
      " \"cocoa_land_used_ha\": \"Land under cocoa (ha)\",\n",
      " \"cocoa_land_used_morethan5_ha\": \"Productive land under cocoa (ha)\",\n",
      " \"cocoa_land_owned_ha\": \"Cocoa land owned (ha)\",\n",
      " \"cocoa_land_leased_ha\": \"Cocoa land leased (ha) \",\n",
      " \"cocoa_land_scrop_abunu_ha\": \"Cocoa land under abunu sharecropping arrangement (ha)\",\n",
      " \"cocoa_land_scrop_abusa_ha\": \"Cocoa land under abusa sharecropping arrangement (ha)\",\n",
      " \"cocoa_land_numlandparcels\": \"Number of cocoa plots\",\n",
      " \"cocoa_trees_num_known_yn\": \"Do you know approximately how many cocoa trees you have?\",\n",
      " \"cocoa_trees_num\": \"Number of cocoa trees\",\n",
      " \"cocoa_years\": \"Years growing cocoa\",\n",
      " \"cocoa_trees_most\": \"Age of most cocoa trees\",\n",
      " \"cocoa_trees_age_5yo_ha\": \"Hectares with trees 5 years or younger\",\n",
      " \"cocoa_trees_age_5_25yo_ha\": \"Hectares with trees between 5 and 25 years old\",\n",
      " \"cocoa_trees_age_25yo_ha\": \"Hectares with trees older than 25 years old\",\n",
      " \"cocoa_land_incdecnc_5ago\": \"Compared with 5 years ago, household increased, decreased, kept the same land\",\n",
      " \"cocoa_land_inc_how_5ago\": \"How did your household increase the amount of land under cocoa?\",\n",
      " \"cocoa_land_dec_how_5ago\": \"How did your household decrease the amount of land under cocoa?\",\n",
      " \"cocoa_intercrop_yn\": \"Cocoa is intercropped\",\n",
      " \"cocoa_intercrop_5yo\": \"Intercropping (with cocoa trees 5 years or younger)\",\n",
      " \"cocoa_intercrop_5_25yo\": \"Intercropping (with cocoa trees between 5 years old and 25 years old)\",\n",
      " \"cocoa_intercrop_25yo\": \"Intercropping (with cocoa trees older than 25 years old)\",\n",
      " \"cocoa_intercrop_othercrop_1\": \"Cocoa was intercropped with avocado\",\n",
      " \"cocoa_intercrop_othercrop_2\": \"Cocoa was intercropped with bananas\",\n",
      " \"cocoa_intercrop_othercrop_3\": \"Cocoa was intercropped with soya beans\",\n",
      " \"cocoa_intercrop_othercrop_4\": \"Cocoa was intercropped with beans\",\n",
      " \"cocoa_intercrop_othercrop_5\": \"Cocoa was intercropped with cabbages\",\n",
      " \"cocoa_intercrop_othercrop_6\": \"Cocoa was intercropped with carrots\",\n",
      " \"cocoa_intercrop_othercrop_7\": \"Cocoa was intercropped with cashews\",\n",
      " \"cocoa_intercrop_othercrop_8\": \"Cocoa was intercropped with cassava\",\n",
      " \"cocoa_intercrop_othercrop_9\": \"Cocoa was intercropped with chili\",\n",
      " \"cocoa_intercrop_othercrop_10\": \"Cocoa was intercropped with cocoa\",\n",
      " \"cocoa_intercrop_othercrop_11\": \"Cocoa was intercropped with coconut\",\n",
      " \"cocoa_intercrop_othercrop_12\": \"Cocoa was intercropped with cocoyam\",\n",
      " \"cocoa_intercrop_othercrop_13\": \"Cocoa was intercropped with coffee\",\n",
      " \"cocoa_intercrop_othercrop_14\": \"Cocoa was intercropped with courgette\",\n",
      " \"cocoa_intercrop_othercrop_15\": \"Cocoa was intercropped with cotton\",\n",
      " \"cocoa_intercrop_othercrop_16\": \"Cocoa was intercropped with cow pea\",\n",
      " \"cocoa_intercrop_othercrop_17\": \"Cocoa was intercropped with eggplant\",\n",
      " \"cocoa_intercrop_othercrop_18\": \"Cocoa was intercropped with fodder\",\n",
      " \"cocoa_intercrop_othercrop_19\": \"Cocoa was intercropped with garlic\",\n",
      " \"cocoa_intercrop_othercrop_20\": \"Cocoa was intercropped with ginger\",\n",
      " \"cocoa_intercrop_othercrop_21\": \"Cocoa was intercropped with groundnuts\",\n",
      " \"cocoa_intercrop_othercrop_22\": \"Cocoa was intercropped with lemon\",\n",
      " \"cocoa_intercrop_othercrop_23\": \"Cocoa was intercropped with lime\",\n",
      " \"cocoa_intercrop_othercrop_24\": \"Cocoa was intercropped with maize\",\n",
      " \"cocoa_intercrop_othercrop_25\": \"Cocoa was intercropped with mango\",\n",
      " \"cocoa_intercrop_othercrop_26\": \"Cocoa was intercropped with millet\",\n",
      " \"cocoa_intercrop_othercrop_27\": \"Cocoa was intercropped with okra\",\n",
      " \"cocoa_intercrop_othercrop_28\": \"Cocoa was intercropped with onions\",\n",
      " \"cocoa_intercrop_othercrop_29\": \"Cocoa was intercropped with oranges\",\n",
      " \"cocoa_intercrop_othercrop_30\": \"Cocoa was intercropped with palm\",\n",
      " \"cocoa_intercrop_othercrop_31\": \"Cocoa was intercropped with papaya\",\n",
      " \"cocoa_intercrop_othercrop_32\": \"Cocoa was intercropped with peas\",\n",
      " \"cocoa_intercrop_othercrop_33\": \"Cocoa was intercropped with peppers\",\n",
      " \"cocoa_intercrop_othercrop_34\": \"Cocoa was intercropped with pineapple\",\n",
      " \"cocoa_intercrop_othercrop_35\": \"Cocoa was intercropped with plantain\",\n",
      " \"cocoa_intercrop_othercrop_36\": \"Cocoa was intercropped with irish potato\",\n",
      " \"cocoa_intercrop_othercrop_37\": \"Cocoa was intercropped with sweet potato\",\n",
      " \"cocoa_intercrop_othercrop_38\": \"Cocoa was intercropped with rice\",\n",
      " \"cocoa_intercrop_othercrop_39\": \"Cocoa was intercropped with rubber\",\n",
      " \"cocoa_intercrop_othercrop_40\": \"Cocoa was intercropped with spinach\",\n",
      " \"cocoa_intercrop_othercrop_41\": \"Cocoa was intercropped with sorghum\",\n",
      " \"cocoa_intercrop_othercrop_42\": \"Cocoa was intercropped with shea nut\",\n",
      " \"cocoa_intercrop_othercrop_43\": \"Cocoa was intercropped with sugar cane\",\n",
      " \"cocoa_intercrop_othercrop_44\": \"Cocoa was intercropped with tangerine\",\n",
      " \"cocoa_intercrop_othercrop_45\": \"Cocoa was intercropped with tomatoes\",\n",
      " \"cocoa_intercrop_othercrop_46\": \"Cocoa was intercropped with watermelon\",\n",
      " \"cocoa_intercrop_othercrop_47\": \"Cocoa was intercropped with yam\",\n",
      " \"cocoa_intercrop_othercrop_48\": \"Cocoa was intercropped with other\",\n",
      " \"cocoa_intercrop_othercrop_49\": \"Cocoa was intercropped with none/ not applicable\",\n",
      " \"cocoa_season1_start\": \"Start of the main cocoa season\",\n",
      " \"cocoa_season1_end\": \"End of the main cocoa season\",\n",
      " \"cocoa_season2_start\": \"Start of the light cocoa season\",\n",
      " \"cocoa_season2_end\": \"End of the light cocoa season\",\n",
      " \"cocoa_prod_known_yn\": \"Do you know how much cocoa your household produced?\",\n",
      " \"cocoa_prod_season1_kg\": \"Cocoa production in the main season (kg)\",\n",
      " \"cocoa_prod_season1_kgha\": \"Cocoa production in the main season (kg/ha)\",\n",
      " \"cocoa_prod_season2_kg\": \"Cocoa production in the light season (kg)\",\n",
      " \"cocoa_prod_season2_kgha\": \"Cocoa production in the light season (kg/ha)\",\n",
      " \"cocoa_prod_total_kgs\": \"Cocoa production (kg)\",\n",
      " \"cocoa_prod_total_kgsha\": \"Cocoa production (kg/ha)\",\n",
      " \"cocoa_valueproduction_usd\": \"Value of cocoa production (usd)\",\n",
      " \"cocoa_valueproduction_usd_ha\": \"Value of cocoa production (usd/ha)\",\n",
      " \"cocoa_profit1_usdha\": \"Revenues minus inputs (usd/ha)\",\n",
      " \"cocoa_profit2_usdha\": \"Revenues minus inputs (usd/ha) minus hired labour (usd/ha)\",\n",
      " \"cocoa_landclear_hi_n_ha\": \"Cocoa land clearing hired labour (days/ha)\",\n",
      " \"fm_cocoa_landclear_hi_n_ha\": \"Cocoa land clearing hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_landclear_hh_n_ha\": \"Cocoa land clearing household labour (days/ha)\",\n",
      " \"fm_cocoa_landclear_hh_n_ha\": \"Cocoa land clearing household labour (days/ha), with inputed values\",\n",
      " \"cocoa_landclear_com_n_ha\": \"Cocoa land clearing communal labour (days/ha)\",\n",
      " \"fm_cocoa_landclear_com_n_ha\": \"Cocoa land clearing communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_landclear_lab_n_ha\": \"Cocoa land clearing total labour total labour (days/ha)\",\n",
      " \"fm_cocoa_landclear_lab_n_ha\": \"Cocoa land clearing total labour total labour (days/ha), with inputed values\",\n",
      " \"cocoa_landclear_hi_lab_usdha\": \"Cocoa land clearing hired labour (usd/ha)\",\n",
      " \"fm_cocoa_landclear_hi_lab_usdha\": \"Cocoa land clearing hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_landprep_hi_n_ha\": \"Cocoa land preparation hired labour (days/ha)\",\n",
      " \"fm_cocoa_landprep_hi_n_ha\": \"Cocoa land preparation hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_landprep_hh_n_ha\": \"Cocoa land preparation household labour (days/ha)\",\n",
      " \"fm_cocoa_landprep_hh_n_ha\": \"Cocoa land preparation household labour (days/ha), with inputed values\",\n",
      " \"cocoa_landprep_com_n_ha\": \"Cocoa land preparation communal labour (days/ha)\",\n",
      " \"fm_cocoa_landprep_com_n_ha\": \"Cocoa land preparation communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_landprep_lab_n_ha\": \"Cocoa land preparation total labour (days/ha)\",\n",
      " \"fm_cocoa_landprep_lab_n_ha\": \"Cocoa land preparation total labour (days/ha), with inputed values\",\n",
      " \"cocoa_landprep_hi_lab_usdha\": \"Cocoa land preparation hired labour (usd/ha)\",\n",
      " \"fm_cocoa_landprep_hi_lab_usdha\": \"Cocoa land preparation hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_planting_hi_n_ha\": \"Cocoa planting hired labour (days/ha)\",\n",
      " \"fm_cocoa_planting_hi_n_ha\": \"Cocoa planting hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_planting_hh_n_ha\": \"Cocoa planting household labour (days/ha)\",\n",
      " \"fm_cocoa_planting_hh_n_ha\": \"Cocoa planting household labour (days/ha), with inputed values\",\n",
      " \"cocoa_planting_com_n_ha\": \"Cocoa planting communal labour (days/ha)\",\n",
      " \"fm_cocoa_planting_com_n_ha\": \"Cocoa planting communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_planting_lab_n_ha\": \"Cocoa planting total labour (days/ha)\",\n",
      " \"fm_cocoa_planting_lab_n_ha\": \"Cocoa planting total labour (days/ha), with inputed values\",\n",
      " \"cocoa_planting_hi_lab_usdha\": \"Cocoa planting hired labour (usd/ha)\",\n",
      " \"fm_cocoa_planting_hi_lab_usdha\": \"Cocoa planting hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_gfert_hi_n_ha\": \"Cocoa granular fertilizer application hired labour (days/ha)\",\n",
      " \"fm_cocoa_gfert_hi_n_ha\": \"Cocoa granular fertilizer application hired labour (days/ha), with inputed value\",\n",
      " \"cocoa_gfert_hh_n_ha\": \"Cocoa granular fertilizer application household labour (days/ha)\",\n",
      " \"fm_cocoa_gfert_hh_n_ha\": \"Cocoa granular fertilizer application household labour (days/ha), with inputed v\",\n",
      " \"cocoa_gfert_com_n_ha\": \"Cocoa granular fertilizer application communal labour (days/ha)\",\n",
      " \"fm_cocoa_gfert_com_n_ha\": \"Cocoa granular fertilizer application communal labour (days/ha), with inputed va\",\n",
      " \"cocoa_gfert_lab_n_ha\": \"Cocoa granular fertilizer application total labour (days/ha)\",\n",
      " \"fm_cocoa_gfert_lab_n_ha\": \"Cocoa granular fertilizer application total labour (days/ha), with inputed value\",\n",
      " \"cocoa_gfert_hi_lab_usdha\": \"Cocoa granular fertilizer application hired labour (usd/ha)\",\n",
      " \"fm_cocoa_gfert_hi_lab_usdha\": \"Cocoa granular fertilizer application hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_lfert_hi_n_ha\": \"Cocoa liquid fertilizer application hired labour (days/ha)\",\n",
      " \"fm_cocoa_lfert_hi_n_ha\": \"Cocoa liquid fertilizer application hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_lfert_hh_n_ha\": \"Cocoa liquid fertilizer application household labour (days/ha)\",\n",
      " \"fm_cocoa_lfert_hh_n_ha\": \"Cocoa liquid fertilizer application household labour (days/ha), with inputed val\",\n",
      " \"cocoa_lfert_com_n_ha\": \"Cocoa liquid fertilizer application communal labour (days/ha)\",\n",
      " \"fm_cocoa_lfert_com_n_ha\": \"Cocoa liquid fertilizer application communal labour (days/ha), with inputed valu\",\n",
      " \"cocoa_lfert_lab_n_ha\": \"Cocoa liquid fertilizer application total labour (days/ha)\",\n",
      " \"fm_cocoa_lfert_lab_n_ha\": \"Cocoa liquid fertilizer application total labour (days/ha), with inputed values\",\n",
      " \"cocoa_lfert_hi_lab_usdha\": \"Cocoa liquid fertilizer application hired labour (usd/ha)\",\n",
      " \"fm_cocoa_lfert_hi_lab_usdha\": \"Cocoa liquid fertilizer application hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_manurecomp_hi_n_ha\": \"Cocoa manure or compost hired labour (days/ha)\",\n",
      " \"fm_cocoa_manurecomp_hi_n_ha\": \"Cocoa manure or compost hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_manurecomp_hh_n_ha\": \"Cocoa manure or compost household labour (days/ha)\",\n",
      " \"fm_cocoa_manurecomp_hh_n_ha\": \"Cocoa manure or compost household labour (days/ha), with inputed values\",\n",
      " \"cocoa_manurecomp_com_n_ha\": \"Cocoa manure or compost communal labour (days/ha)\",\n",
      " \"fm_cocoa_manurecomp_com_n_ha\": \"Cocoa manure or compost communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_manurecomp_lab_n_ha\": \"Cocoa manure or compost total labour (days/ha)\",\n",
      " \"fm_cocoa_manurecomp_lab_n_ha\": \"Cocoa manure or compost total labour (days/ha), with inputed values\",\n",
      " \"cocoa_manurecomp_hi_lab_usdha\": \"Cocoa manure or compost hired labour (usd/ha)\",\n",
      " \"fm_cocoa_manurecomp_hi_lab_usdha\": \"Cocoa manure or compost hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_herb_hi_n_ha\": \"Cocoa herbicide application hired labour (days/ha)\",\n",
      " \"fm_cocoa_herb_hi_n_ha\": \"Cocoa herbicide application hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_herb_hh_n_ha\": \"Cocoa herbicide application household labour (days/ha)\",\n",
      " \"fm_cocoa_herb_hh_n_ha\": \"Cocoa herbicide application household labour (days/ha), with inputed values\",\n",
      " \"cocoa_herb_com_n_ha\": \"Cocoa herbicide application communal labour (days/ha)\",\n",
      " \"fm_cocoa_herb_com_n_ha\": \"Cocoa herbicide application communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_herb_lab_n_ha\": \"Cocoa herbicide application total labour (days/ha)\",\n",
      " \"fm_cocoa_herb_lab_n_ha\": \"Cocoa herbicide application total labour (days/ha), with inputed values\",\n",
      " \"cocoa_herb_hi_lab_usdha\": \"Cocoa herbicide application hired labour (usd/ha)\",\n",
      " \"fm_cocoa_herb_hi_lab_usdha\": \"Cocoa herbicide application hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_pest_hi_n_ha\": \"Cocoa pesticide application hired labour (days/ha)\",\n",
      " \"fm_cocoa_pest_hi_n_ha\": \"Cocoa pesticide application hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_pest_hh_n_ha\": \"Cocoa pesticide application household labour (days/ha)\",\n",
      " \"fm_cocoa_pest_hh_n_ha\": \"Cocoa pesticide application household labour (days/ha), with inputed values\",\n",
      " \"cocoa_pest_com_n_ha\": \"Cocoa pesticide application communal labour (days/ha)\",\n",
      " \"fm_cocoa_pest_com_n_ha\": \"Cocoa pesticide application communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_pest_lab_n_ha\": \"Cocoa pesticide application total labour (days/ha)\",\n",
      " \"fm_cocoa_pest_lab_n_ha\": \"Cocoa pesticide application total labour (days/ha), with inputed values\",\n",
      " \"cocoa_pest_hi_lab_usdha\": \"Cocoa pesticide application hired labour (usd/ha)\",\n",
      " \"fm_cocoa_pest_hi_lab_usdha\": \"Cocoa pesticide application hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_fung_hi_n_ha\": \"Cocoa fungicide application hired labour (days/ha)\",\n",
      " \"fm_cocoa_fung_hi_n_ha\": \"Cocoa fungicide application hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_fung_hh_n_ha\": \"Cocoa fungicide application household labour (days/ha)\",\n",
      " \"fm_cocoa_fung_hh_n_ha\": \"Cocoa fungicide application household labour (days/ha), with inputed values\",\n",
      " \"cocoa_fung_com_n_ha\": \"Cocoa fungicide application communal labour (days/ha)\",\n",
      " \"fm_cocoa_fung_com_n_ha\": \"Cocoa fungicide application communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_fung_lab_n_ha\": \"Cocoa fungicide application total labour (days/ha)\",\n",
      " \"fm_cocoa_fung_lab_n_ha\": \"Cocoa fungicide application total labour (days/ha), with inputed values\",\n",
      " \"cocoa_fung_hi_lab_usdha\": \"Cocoa fungicide application hired labour (usd/ha)\",\n",
      " \"fm_cocoa_fung_hi_lab_usdha\": \"Cocoa fungicide application hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_weed_hi_n_ha\": \"Cocoa weeding hired labour (days/ha)\",\n",
      " \"fm_cocoa_weed_hi_n_ha\": \"Cocoa weeding hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_weed_hh_n_ha\": \"Cocoa weeding household labour (days/ha)\",\n",
      " \"fm_cocoa_weed_hh_n_ha\": \"Cocoa weeding household labour (days/ha), with inputed values\",\n",
      " \"cocoa_weed_com_n_ha\": \"Cocoa weeding communal labour (days/ha)\",\n",
      " \"fm_cocoa_weed_com_n_ha\": \"Cocoa weeding communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_weed_lab_n_ha\": \"Cocoa weeding total labour (days/ha)\",\n",
      " \"fm_cocoa_weed_lab_n_ha\": \"Cocoa weeding total labour (days/ha), with inputed values\",\n",
      " \"cocoa_weed_hi_lab_usdha\": \"Cocoa weeding hired labour (usd/ha)\",\n",
      " \"fm_cocoa_weed_hi_lab_usdha\": \"Cocoa weeding hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_pruning_hi_n_ha\": \"Cocoa pruning hired labour (days/ha)\",\n",
      " \"fm_cocoa_pruning_hi_n_ha\": \"Cocoa pruning hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_pruning_hh_n_ha\": \"Cocoa pruning household labour (days/ha)\",\n",
      " \"fm_cocoa_pruning_hh_n_ha\": \"Cocoa pruning household labour (days/ha), with inputed values\",\n",
      " \"cocoa_pruning_com_n_ha\": \"Cocoa pruning communal labour (days/ha)\",\n",
      " \"fm_cocoa_pruning_com_n_ha\": \"Cocoa pruning communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_pruning_lab_n_ha\": \"Cocoa pruning total labour (days/ha)\",\n",
      " \"fm_cocoa_pruning_lab_n_ha\": \"Cocoa pruning total labour (days/ha), with inputed values\",\n",
      " \"cocoa_pruning_hi_lab_usdha\": \"Cocoa pruning hired labour (usd/ha)\",\n",
      " \"fm_cocoa_pruning_hi_lab_usdha\": \"Cocoa pruning hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_harvest_hi_n_ha\": \"Cocoa harvesting hired labour (days/ha)\",\n",
      " \"fm_cocoa_harvest_hi_n_ha\": \"Cocoa harvesting hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_harvest_hh_n_ha\": \"Cocoa harvesting household labour (days/ha)\",\n",
      " \"fm_cocoa_harvest_hh_n_ha\": \"Cocoa harvesting household labour (days/ha), with inputed values\",\n",
      " \"cocoa_harvest_com_n_ha\": \"Cocoa harvesting communal labour (days/ha)\",\n",
      " \"fm_cocoa_harvest_com_n_ha\": \"Cocoa harvesting communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_harvest_lab_n_ha\": \"Cocoa harvesting total labour (days/ha)\",\n",
      " \"fm_cocoa_harvest_lab_n_ha\": \"Cocoa harvesting total labour (days/ha), with inputed values\",\n",
      " \"cocoa_harvest_hi_lab_usdha\": \"Cocoa harvesting hired labour (usd/ha)\",\n",
      " \"fm_cocoa_harvest_hi_lab_usdha\": \"Cocoa harvesting hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_podbreak_hi_n_ha\": \"Cocoa pod breaking hired labour (days/ha)\",\n",
      " \"fm_cocoa_podbreak_hi_n_ha\": \"Cocoa pod breaking hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_podbreak_hh_n_ha\": \"Cocoa pod breaking household labour (days/ha)\",\n",
      " \"fm_cocoa_podbreak_hh_n_ha\": \"Cocoa pod breaking household labour (days/ha), with inputed values\",\n",
      " \"cocoa_podbreak_com_n_ha\": \"Cocoa pod breaking communal labour (days/ha)\",\n",
      " \"fm_cocoa_podbreak_com_n_ha\": \"Cocoa pod breaking communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_podbreak_lab_n_ha\": \"Cocoa pod breaking total labour (days/ha)\",\n",
      " \"fm_cocoa_podbreak_lab_n_ha\": \"Cocoa pod breaking total labour (days/ha), with inputed values\",\n",
      " \"cocoa_podbreak_hi_lab_usdha\": \"Cocoa pod breaking hired labour (usd/ha)\",\n",
      " \"fm_cocoa_podbreak_hi_lab_usdha\": \"Cocoa pod breaking hired labour (usd/ha), with inputed values\",\n",
      " \"cocoa_transport_hi_n_ha\": \"Cocoa transporting hired labour (days/ha)\",\n",
      " \"fm_cocoa_transport_hi_n_ha\": \"Cocoa transporting hired labour (days/ha), with inputed values\",\n",
      " \"cocoa_transport_hh_n_ha\": \"Cocoa transporting household labour (days/ha)\",\n",
      " \"fm_cocoa_transport_hh_n_ha\": \"Cocoa transporting household labour (days/ha), with inputed values\",\n",
      " \"cocoa_transport_com_n_ha\": \"Cocoa transporting communal labour (days/ha)\",\n",
      " \"fm_cocoa_transport_com_n_ha\": \"Cocoa transporting communal labour (days/ha), with inputed values\",\n",
      " \"cocoa_transport_lab_n_ha\": \"Cocoa transporting total labour (days/ha)\",\n",
      " \"fm_cocoa_transport_lab_n_ha\": \"Cocoa transporting total labour (days/ha), with inputed values\",\n",
      " \"cocoa_transport_hi_lab_usdha\": \"Cocoa transporting hired labour (usd/ha)\",\n",
      " \"fm_cocoa_transport_hi_lab_usdha\": \"Cocoa transporting hired labour (usd/ha), with inputed values\",\n",
      " \"fm_cocoa_hh_n_ha\": \"Total cocoa household labour (days/ha), with inputed values\",\n",
      " \"fm_cocoa_hi_n_ha\": \"Total cocoa hired labour (days/ha), with inputed values\",\n",
      " \"fm_cocoa_com_n_ha\": \"Total cocoa communal labour (days/ha), with inputed values\",\n",
      " \"fm_cocoa_lab_n_ha\": \"Total cocoa labour (days/ha), with inputed values\",\n",
      " \"fm_cocoa_hi_lab_usdha\": \"Cocoa hired labour costs (usd/ha), with inputed values\",\n",
      " \"fm_cocoa_inputs_usdha\": \"Cocoa input costs (usd/ha), with inputed values\",\n",
      " \"cocoa_varieties_1\": \"Cocoa variety: tetteh quarshie\",\n",
      " \"cocoa_varieties_2\": \"Cocoa variety: amazonia\",\n",
      " \"cocoa_varieties_3\": \"Cocoa variety: hybrid\",\n",
      " \"cocoa_varieties_7\": \"Cocoa variety: hybride (2nd generation)\",\n",
      " \"cocoa_varieties_4\": \"Cocoa variety: other\",\n",
      " \"cocoa_varieties_5\": \"Cocoa variety: french\",\n",
      " \"cocoa_varieties_6\": \"Cocoa variety: tout venant\",\n",
      " \"cocoa_varieties_999\": \"Cocoa variety: don't know\",\n",
      " \"cocoa_trees_rows_yn\": \"Trees are planted in rows\",\n",
      " \"cocoa_trees_distance_known_yn\": \"Do you know the average distance between cocoa trees on your household's farm?\",\n",
      " \"cocoa_tree_spacing_trees\": \"Average distance between your household's cocoa trees\",\n",
      " \"cocoa_tree_spacing_rows\": \"Average distance between rows\",\n",
      " \"cocoa_landclear_yn\": \"Land clearing for cocoa was done last year\",\n",
      " \"cocoa_landclear_times\": \"Number of times land was cleared for cocoa last year\",\n",
      " \"cocoa_landclear_ha\": \"Land cleared for cocoa last year (ha)\",\n",
      " \"cocoa_landclear_hhhiredcom_1\": \"Land clearing done by household labour\",\n",
      " \"cocoa_landclear_hhhiredcom_2\": \"Land clearing done by hired labour\",\n",
      " \"cocoa_landclear_hhhiredcom_3\": \"Land clearing done by communal labour\",\n",
      " \"cocoa_landclear_hhhiredcom_4\": \"Land clearing done by company/codapec labour\",\n",
      " \"cocoa_landclear_hh_hi\": \"Land clearing was done by household, hired labour or both\",\n",
      " \"cocoa_landclear_gender_1\": \"Land clearing was done by man\",\n",
      " \"cocoa_landclear_gender_2\": \"Land clearing was done by women\",\n",
      " \"cocoa_landclear_gender\": \"Land clearing was done by man, women or both\",\n",
      " \"cocoa_landclear_daysknown_yn\": \"Do you know how many labour days were spent doing land clearing for cocoa?\",\n",
      " \"cocoa_landclear_orig_hi_cost\": \"Daily rate for a hired labourer for land clearing (lcu)\",\n",
      " \"cocoa_landclear_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for land clearing (lc\",\n",
      " \"cocoa_landprep_yn\": \"Land preparation for cocoa was done last year\",\n",
      " \"cocoa_landprep_times\": \"Number of times land was prepared for cocoa last year\",\n",
      " \"cocoa_landprep_ha\": \"Land prepared for cocoa last year (ha)\",\n",
      " \"cocoa_landprep_how_1\": \"Land preparation: manual labour\",\n",
      " \"cocoa_landprep_how_2\": \"Land preparation: animal plough\",\n",
      " \"cocoa_landprep_how_3\": \"Land preparation: tractor\",\n",
      " \"cocoa_landprep_how_4\": \"Land preparation: other\",\n",
      " \"cocoa_landprep_hhhiredcom_1\": \"Land preparation done by household labour\",\n",
      " \"cocoa_landprep_hhhiredcom_2\": \"Land preparation done by hired labour\",\n",
      " \"cocoa_landprep_hhhiredcom_3\": \"Land preparation done by communal labour\",\n",
      " \"cocoa_landprep_hhhiredcom_4\": \"Land preparation done by company/codapec labour\",\n",
      " \"cocoa_landprep_hh_hi\": \"Land preparation was done by household, hired labour or both\",\n",
      " \"cocoa_landprep_gender_1\": \"Land preparation was done by man\",\n",
      " \"cocoa_landprep_gender_2\": \"Land preparation was done by women\",\n",
      " \"cocoa_landprep_gender\": \"Land preparation was done by man, women or both\",\n",
      " \"cocoa_landprep_daysknown_yn\": \"Do you know how many labour days were spent doing land preparation for cocoa?\",\n",
      " \"cocoa_landprep_orig_hi_cost\": \"Daily rate for a hired labourer for land preparation (lcu)\",\n",
      " \"cocoa_landprep_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for land preparation \",\n",
      " \"cocoa_planting_yn\": \"Cocoa planting was done last year\",\n",
      " \"cocoa_planting_times\": \"Number of times cocoa was planted last year\",\n",
      " \"cocoa_planting_ha\": \"Cocoa land planted last year (ha)\",\n",
      " \"cocoa_planting_hhhiredcom_1\": \"Planting done by household labour\",\n",
      " \"cocoa_planting_hhhiredcom_2\": \"Planting done by hired labour\",\n",
      " \"cocoa_planting_hhhiredcom_3\": \"Planting done by communal labour\",\n",
      " \"cocoa_planting_hhhiredcom_4\": \"Planting done by company/codapec labour\",\n",
      " \"cocoa_planting_hh_hi\": \"Planting was done by household, hired labour or both\",\n",
      " \"cocoa_planting_gender_1\": \"Planting was done by man\",\n",
      " \"cocoa_planting_gender_2\": \"Planting was done by women\",\n",
      " \"cocoa_planting_gender\": \"Planting was done by man, women or both\",\n",
      " \"cocoa_planting_daysknown_yn\": \"Do you know how many labour days were spent doing planting for cocoa?\",\n",
      " \"cocoa_planting_orig_hi_cost\": \"Daily rate for a hired labourer for planting (lcu)\",\n",
      " \"cocoa_planting_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for planting (lcu)\",\n",
      " \"cocoa_planting_seed_origin_1\": \"Seeds source: own farm\",\n",
      " \"cocoa_planting_seed_origin_2\": \"Seeds source: neighbours / friends / relatives farms\",\n",
      " \"cocoa_planting_seed_origin_3\": \"Seeds source: farmer organisation\",\n",
      " \"cocoa_planting_seed_origin_4\": \"Seeds source: government, extension\",\n",
      " \"cocoa_planting_seed_origin_5\": \"Seeds source: purchased from agrodealer shop\",\n",
      " \"cocoa_planting_seed_origin_6\": \"Seeds source: trader\",\n",
      " \"cocoa_planting_seed_origin_7\": \"Seeds source: company / lbc\",\n",
      " \"cocoa_planting_seed_origin_8\": \"Seeds source: other\",\n",
      " \"cocoa_planting_seed_origin_9\": \"Seeds source: nestle, cargill, olam, ecom other company\",\n",
      " \"cocoa_planting_seed_origin_10\": \"Seeds source: ong\",\n",
      " \"cocoa_planting_seed_origin_999\": \"Seeds source: don't know\",\n",
      " \"cocoa_gfert_yn\": \"Granular fertilizer was applied on cocoa land\",\n",
      " \"cocoa_gfert_times\": \"Number of times granular fertilizer was applied on cocoa last year\",\n",
      " \"cocoa_gfert_given_yn\": \"Granular fertilizer given\",\n",
      " \"cocoa_gfert_purchased_yn\": \"Granular fertilizer bought\",\n",
      " \"cocoa_gfert_given_pack\": \"Number of granular fertizer bags given\",\n",
      " \"cocoa_gfert_purchased_pack\": \"Number of granular fertizer bags bought\",\n",
      " \"cocoa_gfert_pack_used_total\": \"Number of granular fertizer bags applied\",\n",
      " \"cocoa_gfert_given_unitha\": \"Granular fertilizer (kg,l/ha given)\",\n",
      " \"cocoa_gfert_purchased_unitha\": \"Granular fertilizer (kg,l/ha purchased)\",\n",
      " \"cocoa_gfert_applied_unitha\": \"Granular fertilizer (kg,l/ha appllied)\",\n",
      " \"cocoa_gfert_usdha\": \"Granular fertilizer expenses (usd/ha)\",\n",
      " \"fm_cocoa_gfert_usdha\": \"Granular fertilizer (usd/ha), with inputed values\",\n",
      " \"cocoa_gfert_hhhiredcom_1\": \"Granular fertilizer application done by household labour\",\n",
      " \"cocoa_gfert_hhhiredcom_2\": \"Granular fertilizer application done by hired labour\",\n",
      " \"cocoa_gfert_hhhiredcom_3\": \"Granular fertilizer application done by communal labour\",\n",
      " \"cocoa_gfert_hhhiredcom_4\": \"Granular fertilizer application done by company/codapec labour\",\n",
      " \"cocoa_gfert_hh_hi\": \"Granular fertilizer application was done by household, hired labour or both\",\n",
      " \"cocoa_gfert_gender_1\": \"Granular fertilizer application was done by man\",\n",
      " \"cocoa_gfert_gender_2\": \"Granular fertilizer application was done by women\",\n",
      " \"cocoa_gfert_gender\": \"Granular fertilizer application was done by man, women or both\",\n",
      " \"cocoa_gfert_daysknown_yn\": \"Do you know how many labour days were spent doing fertilizer application?\",\n",
      " \"cocoa_gfert_orig_hi_cost\": \"Daily rate for a hired labourer for granular fertilizer application (lcu)\",\n",
      " \"cocoa_gfert_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for granular fertiliz\",\n",
      " \"cocoa_lfert_yn\": \"Liquid fertilizer was applied on cocoa land\",\n",
      " \"cocoa_lfert_times\": \"Number of times liquid fertilizer was applied on cocoa last year\",\n",
      " \"cocoa_lfert_given_yn\": \"Liquid fertilizer given\",\n",
      " \"cocoa_lfert_purchased_yn\": \"Liquid fertilizer bought\",\n",
      " \"cocoa_lfert_given_pack\": \"Number of liquid fertizer bags given\",\n",
      " \"cocoa_lfert_purchased_pack\": \"Number of liquid fertizer bags bought\",\n",
      " \"cocoa_lfert_pack_used_total\": \"Number of liquid fertizer bags applied\",\n",
      " \"cocoa_lfert_given_unitha\": \"Liquid fertilizer (kg,l/ha given)\",\n",
      " \"cocoa_lfert_purchased_unitha\": \"Liquid fertilizer (kg,l/ha purchased)\",\n",
      " \"cocoa_lfert_applied_unitha\": \"Liquid fertilizer (kg,l/ha appllied)\",\n",
      " \"cocoa_lfert_usdha\": \"Liquid fertilizer expenses (usd/ha)\",\n",
      " \"fm_cocoa_lfert_usdha\": \"Liquid fertilizer (usd/ha), with inputed values\",\n",
      " \"cocoa_lfert_hhhiredcom_1\": \"Liquid fertilizer application done by household labour\",\n",
      " \"cocoa_lfert_hhhiredcom_2\": \"Liquid fertilizer application done by hired labour\",\n",
      " \"cocoa_lfert_hhhiredcom_3\": \"Liquid fertilizer application done by communal labour\",\n",
      " \"cocoa_lfert_hhhiredcom_4\": \"Liquid fertilizer application done by company/codapec labour\",\n",
      " \"cocoa_lfert_hh_hi\": \"Liquid fertilizer application was done by household, hired labour or both\",\n",
      " \"cocoa_lfert_gender_1\": \"Liquid fertilizer application was done by man\",\n",
      " \"cocoa_lfert_gender_2\": \"Liquid fertilizer application was done by women\",\n",
      " \"cocoa_lfert_gender\": \"Liquid fertilizer application was done by man, women or both\",\n",
      " \"cocoa_lfert_daysknown_yn\": \"Do you know how many labour days were spent doing liquid fertilizer application?\",\n",
      " \"cocoa_lfert_orig_hi_cost\": \"Daily rate for a hired labourer for liquid fertilizer application (lcu)\",\n",
      " \"cocoa_lfert_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for liquid fertilizer\",\n",
      " \"cocoa_manurecomp_yn\": \"Manure or compost was applied on cocoa land\",\n",
      " \"cocoa_manurecomp_times\": \"Number of times manure or compost was applied on cocoa last year\",\n",
      " \"cocoa_manurecomp_hhhiredcom_1\": \"Manure or compost application done by household labour\",\n",
      " \"cocoa_manurecomp_hhhiredcom_2\": \"Manure or compost application done by hired labour\",\n",
      " \"cocoa_manurecomp_hhhiredcom_3\": \"Manure or compost application done by communal labour\",\n",
      " \"cocoa_manurecomp_hhhiredcom_4\": \"Manure or compost application done by company/codapec labour\",\n",
      " \"cocoa_manurecomp_hh_hi\": \"Manure or compost application was done by household, hired labour or both\",\n",
      " \"cocoa_manurecomp_gender_1\": \"Manure or compost application was done by man\",\n",
      " \"cocoa_manurecomp_gender_2\": \"Manure or compost application was done by women\",\n",
      " \"cocoa_manurecomp_gender\": \"Manure or compost application was done by man, women or both\",\n",
      " \"cocoa_manurecomp_daysknown_yn\": \"Do you know how many labour days were spent doing manure or compost application?\",\n",
      " \"cocoa_manurecomp_orig_hi_cost\": \"Daily rate for a hired labourer for manure or compost application (lcu)\",\n",
      " \"cocoa_manurecomp_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for manure or compost\",\n",
      " \"cocoa_herb_yn\": \"Herbicides was applied on cocoa land\",\n",
      " \"cocoa_herb_times\": \"Number of times herbicides was applied on cocoa last year\",\n",
      " \"cocoa_herb_given_yn\": \"Herbicides given\",\n",
      " \"cocoa_herb_purchased_yn\": \"Herbicides bought\",\n",
      " \"cocoa_herb_given_pack\": \"Number of liquid fertizer bags given\",\n",
      " \"cocoa_herb_purchased_pack\": \"Number of liquid fertizer bags bought\",\n",
      " \"cocoa_herb_pack_used_total\": \"Number of liquid fertizer bags applied\",\n",
      " \"cocoa_herb_given_unitha\": \"Herbicide (kg,l/ha given)\",\n",
      " \"cocoa_herb_purchased_unitha\": \"Herbicide (kg,l/ha purchased)\",\n",
      " \"cocoa_herb_applied_unitha\": \"Herbicide (kg,l/ha appllied)\",\n",
      " \"cocoa_herb_usdha\": \"Herbicides expenses (usd/ha)\",\n",
      " \"fm_cocoa_herb_usdha\": \"Herbicide (usd/ha), with inputed values\",\n",
      " \"cocoa_herb_hhhiredcom_1\": \"Herbicide application done by household labour\",\n",
      " \"cocoa_herb_hhhiredcom_2\": \"Herbicide application done by hired labour\",\n",
      " \"cocoa_herb_hhhiredcom_3\": \"Herbicide application done by communal labour\",\n",
      " \"cocoa_herb_hhhiredcom_4\": \"Herbicide application done by company/codapec labour\",\n",
      " \"cocoa_herb_hh_hi\": \"Herbicide application was done by household, hired labour or both\",\n",
      " \"cocoa_herb_gender_1\": \"Herbicide application was done by man\",\n",
      " \"cocoa_herb_gender_2\": \"Herbicide application was done by women\",\n",
      " \"cocoa_herb_gender\": \"Herbicide application was done by man, women or both\",\n",
      " \"cocoa_herb_daysknown_yn\": \"Do you know how many labour days were spent doing herbicide application?\",\n",
      " \"cocoa_herb_orig_hi_cost\": \"Daily rate for a hired labourer for herbicide application (lcu)\",\n",
      " \"cocoa_herb_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for herbicide applica\",\n",
      " \"cocoa_pest_yn\": \"Pesticides was applied on cocoa land\",\n",
      " \"cocoa_pest_times\": \"Number of times pesticides was applied on cocoa last year\",\n",
      " \"cocoa_pest_given_yn\": \"Pesticides given\",\n",
      " \"cocoa_pest_purchased_yn\": \"Pesticides bought\",\n",
      " \"cocoa_pest_given_pack\": \"Number of liquid fertizer bags given\",\n",
      " \"cocoa_pest_purchased_pack\": \"Number of liquid fertizer bags bought\",\n",
      " \"cocoa_pest_pack_used_total\": \"Number of liquid fertizer bags applied\",\n",
      " \"cocoa_pest_given_unitha\": \"Pesticide or insecticide (kg,l/ha given)\",\n",
      " \"cocoa_pest_purchased_unitha\": \"Pesticide or insecticide (kg,l/ha purchased)\",\n",
      " \"cocoa_pest_applied_unitha\": \"Pesticide or insecticide (kg,l/ha appllied)\",\n",
      " \"cocoa_pest_usdha\": \"Pesticides expenses (usd/ha)\",\n",
      " \"fm_cocoa_pest_usdha\": \"Pesticide or insecticide (usd/ha), with inputed values\",\n",
      " \"cocoa_pest_hhhiredcom_1\": \"Pesticide application done by household labour\",\n",
      " \"cocoa_pest_hhhiredcom_2\": \"Pesticide application done by hired labour\",\n",
      " \"cocoa_pest_hhhiredcom_3\": \"Pesticide application done by communal labour\",\n",
      " \"cocoa_pest_hhhiredcom_4\": \"Pesticide application done by company/codapec labour\",\n",
      " \"cocoa_pest_hh_hi\": \"Pesticide application was done by household, hired labour or both\",\n",
      " \"cocoa_pest_gender_1\": \"Pesticide application was done by man\",\n",
      " \"cocoa_pest_gender_2\": \"Pesticide application was done by women\",\n",
      " \"cocoa_pest_gender\": \"Pesticide application was done by man, women or both\",\n",
      " \"cocoa_pest_daysknown_yn\": \"Do you know how many labour days were spent doing pesticide application?\",\n",
      " \"cocoa_pest_orig_hi_cost\": \"Daily rate for a hired labourer for pesticide application (lcu)\",\n",
      " \"cocoa_pest_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for pesticide applica\",\n",
      " \"cocoa_fung_yn\": \"Fungicides was applied on cocoa land\",\n",
      " \"cocoa_fung_times\": \"Number of times fungicides was applied on cocoa last year\",\n",
      " \"cocoa_fung_given_yn\": \"Fungicides given\",\n",
      " \"cocoa_fung_purchased_yn\": \"Fungicides bought\",\n",
      " \"cocoa_fung_given_pack\": \"Number of liquid fertizer bags given\",\n",
      " \"cocoa_fung_purchased_pack\": \"Number of liquid fertizer bags bought\",\n",
      " \"cocoa_fung_pack_used_total\": \"Number of liquid fertizer bags applied\",\n",
      " \"cocoa_fung_given_unitha\": \"Fungicide (kg,l/ha given)\",\n",
      " \"cocoa_fung_purchased_unitha\": \"Fungicide (kg,l/ha purchased)\",\n",
      " \"cocoa_fung_applied_unitha\": \"Fungicide (kg,l/ha appllied)\",\n",
      " \"cocoa_fung_usdha\": \"Fungicides expenses (usd/ha)\",\n",
      " \"fm_cocoa_fung_usdha\": \"Fungicide (usd/ha), with inputed values\",\n",
      " \"cocoa_fung_hhhiredcom_1\": \"Fungicide application done by household labour\",\n",
      " \"cocoa_fung_hhhiredcom_2\": \"Fungicide application done by hired labour\",\n",
      " \"cocoa_fung_hhhiredcom_3\": \"Fungicide application done by communal labour\",\n",
      " \"cocoa_fung_hhhiredcom_4\": \"Fungicide application done by company/codapec labour\",\n",
      " \"cocoa_fung_hh_hi\": \"Fungicide application was done by household, hired labour or both\",\n",
      " \"cocoa_fung_gender_1\": \"Fungicide application was done by man\",\n",
      " \"cocoa_fung_gender_2\": \"Fungicide application was done by women\",\n",
      " \"cocoa_fung_gender\": \"Fungicide application was done by man, women or both\",\n",
      " \"cocoa_fung_daysknown_yn\": \"Do you know how many labour days were spent doing fungicide application?\",\n",
      " \"cocoa_fung_orig_hi_cost\": \"Daily rate for a hired labourer for fungicide application (lcu)\",\n",
      " \"cocoa_fung_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for fungicide applica\",\n",
      " \"cocoa_weed_yn\": \"Weeding was done on cocoa land\",\n",
      " \"cocoa_weed_times\": \"Number of times weeding was done on cocoa last year\",\n",
      " \"cocoa_weed_hhhiredcom_1\": \"Weeding done by household labour\",\n",
      " \"cocoa_weed_hhhiredcom_2\": \"Weeding done by hired labour\",\n",
      " \"cocoa_weed_hhhiredcom_3\": \"Weeding done by communal labour\",\n",
      " \"cocoa_weed_hhhiredcom_4\": \"Weeding done by company/codapec labour\",\n",
      " \"cocoa_weed_hh_hi\": \"Weeding was done by household, hired labour or both\",\n",
      " \"cocoa_weed_gender_1\": \"Weeding was done by man\",\n",
      " \"cocoa_weed_gender_2\": \"Weeding was done by women\",\n",
      " \"cocoa_weed_gender\": \"Weeding was done by man, women or both\",\n",
      " \"cocoa_weed_daysknown_yn\": \"Do you know how many labour days were spent doing weeding for cocoa?\",\n",
      " \"cocoa_weed_orig_hi_cost\": \"Daily rate for a hired labourer for weeding (lcu)\",\n",
      " \"cocoa_weed_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for weeding (lcu)\",\n",
      " \"cocoa_pruning_yn\": \"Pruuning was done on cocoa land\",\n",
      " \"cocoa_pruning_times\": \"Number of times prunning was done on cocoa last year\",\n",
      " \"cocoa_pruning_hhhiredcom_1\": \"Prunning done by household labour\",\n",
      " \"cocoa_pruning_hhhiredcom_2\": \"Prunning done by hired labour\",\n",
      " \"cocoa_pruning_hhhiredcom_3\": \"Prunning done by communal labour\",\n",
      " \"cocoa_pruning_hhhiredcom_4\": \"Prunning done by company/codapec labour\",\n",
      " \"cocoa_pruning_hh_hi\": \"Prunning was done by household, hired labour or both\",\n",
      " \"cocoa_pruning_gender_1\": \"Prunning was done by man\",\n",
      " \"cocoa_pruning_gender_2\": \"Prunning was done by women\",\n",
      " \"cocoa_pruning_gender\": \"Prunning was done by man, women or both\",\n",
      " \"cocoa_pruning_daysknown_yn\": \"Do you know how many labour days were spent doing pruning for cocoa?\",\n",
      " \"cocoa_pruning_orig_hi_cost\": \"Daily rate for a hired labourer for prunning (lcu)\",\n",
      " \"cocoa_pruning_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for prunning (lcu)\",\n",
      " \"cocoa_harvest_yn\": \"Cocoa was harvested last year\",\n",
      " \"cocoa_harvest_times\": \"Number of times cocoa was harvested last year\",\n",
      " \"cocoa_harvest_months_all_1\": \"Harvesting: january\",\n",
      " \"cocoa_harvest_months_all_2\": \"Harvesting: february\",\n",
      " \"cocoa_harvest_months_all_3\": \"Harvesting: march\",\n",
      " \"cocoa_harvest_months_all_4\": \"Harvesting: april\",\n",
      " \"cocoa_harvest_months_all_5\": \"Harvesting: may\",\n",
      " \"cocoa_harvest_months_all_6\": \"Harvesting: june\",\n",
      " \"cocoa_harvest_months_all_7\": \"Harvesting: july\",\n",
      " \"cocoa_harvest_months_all_8\": \"Harvesting: august\",\n",
      " \"cocoa_harvest_months_all_9\": \"Harvesting: september\",\n",
      " \"cocoa_harvest_months_all_10\": \"Harvesting: october\",\n",
      " \"cocoa_harvest_months_all_11\": \"Harvesting: november\",\n",
      " \"cocoa_harvest_months_all_12\": \"Harvesting: december\",\n",
      " \"cocoa_harvest_months_all_999\": \"Harvesting: don't know\",\n",
      " \"cocoa_harvest_months_all_99\": \"Harvesting: none/ not applicable\",\n",
      " \"cocoa_harvest_hhhiredcom_1\": \"Harvesting done by household labour\",\n",
      " \"cocoa_harvest_hhhiredcom_2\": \"Harvesting done by hired labour\",\n",
      " \"cocoa_harvest_hhhiredcom_3\": \"Harvesting done by communal labour\",\n",
      " \"cocoa_harvest_hhhiredcom_4\": \"Harvesting done by company/codapec labour\",\n",
      " \"cocoa_harvest_hh_hi\": \"Harvesting was done by household, hired labour or both\",\n",
      " \"cocoa_harvest_gender_1\": \"Harvesting was done by man\",\n",
      " \"cocoa_harvest_gender_2\": \"Harvesting was done by women\",\n",
      " \"cocoa_harvest_gender\": \"Harvesting was done by man, women or both\",\n",
      " \"cocoa_harvest_daysknown_yn\": \"Do you know how many labour days were spent doing harvesting for cocoa?\",\n",
      " \"cocoa_harvest_orig_hi_cost\": \"Daily rate for a hired labourer for harvesting (lcu)\",\n",
      " \"cocoa_harvest_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for harvesting (lcu)\",\n",
      " \"cocoa_podbreak_yn\": \"Cocoa pod breaking was done last year\",\n",
      " \"cocoa_podbreak_times\": \"Number of times pod breaking was done on cocoa last year\",\n",
      " \"cocoa_podbreak_wait_days\": \"Number of days o wait before breaking the pods after harvesting\",\n",
      " \"cocoa_podbreak_hhhiredcom_1\": \"Pod breaking done by household labour\",\n",
      " \"cocoa_podbreak_hhhiredcom_2\": \"Pod breaking done by hired labour\",\n",
      " \"cocoa_podbreak_hhhiredcom_3\": \"Pod breaking done by communal labour\",\n",
      " \"cocoa_podbreak_hhhiredcom_4\": \"Pod breaking done by company/codapec labour\",\n",
      " \"cocoa_podbreak_hh_hi\": \"Pod breaking was done by household, hired labour or both\",\n",
      " \"cocoa_podbreak_gender_1\": \"Pod breaking was done by man\",\n",
      " \"cocoa_podbreak_gender_2\": \"Pod breaking was done by women\",\n",
      " \"cocoa_podbreak_gender\": \"Pod breaking was done by man, women or both\",\n",
      " \"cocoa_podbreak_daysknown_yn\": \"Do you know how many labour days were spent doing pod breaking  last year?\",\n",
      " \"cocoa_podbreak_orig_hi_cost\": \"Daily rate for a hired labourer for pod breaking (lcu)\",\n",
      " \"cocoa_podbreak_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for pod breaking (lcu\",\n",
      " \"cocoa_ferment_yn\": \"Cocoa fermenting was done last year\",\n",
      " \"cocoa_ferment_wait_days\": \"Number of days cocoa beans were fermented\",\n",
      " \"cocoa_ferment_material_1\": \"Ferment material: rice bag\",\n",
      " \"cocoa_ferment_material_2\": \"Ferment material: banana leafs (in baskets)\",\n",
      " \"cocoa_ferment_material_3\": \"Ferment material: country mats\",\n",
      " \"cocoa_ferment_material_4\": \"Ferment material: fermentation box\",\n",
      " \"cocoa_ferment_material_5\": \"Ferment material: other\",\n",
      " \"cocoa_ferment_material_999\": \"Ferment material: don't know\",\n",
      " \"cocoa_ferment_material_99\": \"Ferment material: not applicable\",\n",
      " \"cocoa_ferment_hhhiredcom_1\": \"Fermenting done by household labour\",\n",
      " \"cocoa_ferment_hhhiredcom_2\": \"Fermenting done by hired labour\",\n",
      " \"cocoa_ferment_hhhiredcom_3\": \"Fermenting done by communal labour\",\n",
      " \"cocoa_ferment_hhhiredcom_4\": \"Fermenting done by company/codapec labour\",\n",
      " \"cocoa_ferment_hh_hi\": \"Fermenting was done by household, hired labour or both\",\n",
      " \"cocoa_ferment_gender_1\": \"Fermenting was done by man\",\n",
      " \"cocoa_ferment_gender_2\": \"Fermenting was done by women\",\n",
      " \"cocoa_ferment_gender\": \"Fermenting was done by man, women or both\",\n",
      " \"cocoa_transport_yn\": \"Cocoa transported from the farm to the homestead\",\n",
      " \"cocoa_transport_means\": \"Cocoa transport means\",\n",
      " \"cocoa_transport_hhhiredcom_1\": \"Transporting done by household labour\",\n",
      " \"cocoa_transport_hhhiredcom_2\": \"Transporting done by hired labour\",\n",
      " \"cocoa_transport_hhhiredcom_3\": \"Transporting done by communal labour\",\n",
      " \"cocoa_transport_hhhiredcom_4\": \"Transporting done by company/codapec labour\",\n",
      " \"cocoa_transport_hh_hi\": \"Transporting was done by household, hired labour or both\",\n",
      " \"cocoa_transport_gender_1\": \"Transporting was done by man\",\n",
      " \"cocoa_transport_gender_2\": \"Transporting was done by women\",\n",
      " \"cocoa_transport_gender\": \"Transporting was done by man, women or both\",\n",
      " \"cocoa_transport_daysknown_yn\": \"Do you know how many labour days were spent transporting cocoa last season?\",\n",
      " \"cocoa_transport_times\": \"Number of times cocoa was transported last year\",\n",
      " \"cocoa_transport_orig_hi_cost\": \"Daily rate for a hired labourer for transporting (lcu)\",\n",
      " \"cocoa_transport_hi_cost\": \"Median daily rate (comparable region) for a hired labourer for transporting (lcu\",\n",
      " \"cocoa_dry_yn\": \"Cocoa was dried last year\",\n",
      " \"cocoa_dry_wait_days\": \"Number of days cocoa beans were dried\",\n",
      " \"cocoa_dry_material\": \"Cocoa drying material\",\n",
      " \"cocoa_dry_hhhiredcom_1\": \"Drying done by household labour\",\n",
      " \"cocoa_dry_hhhiredcom_2\": \"Drying done by hired labour\",\n",
      " \"cocoa_dry_hhhiredcom_3\": \"Drying done by communal labour\",\n",
      " \"cocoa_dry_hhhiredcom_4\": \"Drying done by company/codapec labour\",\n",
      " \"cocoa_dry_gender_1\": \"Drying was done by man\",\n",
      " \"cocoa_dry_gender_2\": \"Drying was done by women\",\n",
      " \"cocoa_irrigation_yn\": \"Cocoa was irrigated last season\",\n",
      " \"cocoa_irrigation_source_1\": \"Cocoa irrigation source: stream/ river\",\n",
      " \"cocoa_irrigation_source_2\": \"Cocoa irrigation source: lake/pond\",\n",
      " \"cocoa_irrigation_source_3\": \"Cocoa irrigation source: water pan / water reservoir (dug, and maybe lined with \",\n",
      " \"cocoa_irrigation_source_4\": \"Cocoa irrigation source: ground water tank (concrete)\",\n",
      " \"cocoa_irrigation_source_5\": \"Cocoa irrigation source: roof harvesting of rainwater\",\n",
      " \"cocoa_irrigation_source_6\": \"Cocoa irrigation source: borehole/ well\",\n",
      " \"cocoa_irrigation_source_7\": \"Cocoa irrigation source: other\",\n",
      " \"cocoa_irrigation_method_1\": \"Cocoa irrigation method: buckets /watering cans/ bottles (by hand)\",\n",
      " \"cocoa_irrigation_method_2\": \"Cocoa irrigation method: pedal pump\",\n",
      " \"cocoa_irrigation_method_3\": \"Cocoa irrigation method: motorised pump\",\n",
      " \"cocoa_irrigation_method_4\": \"Cocoa irrigation method: drip irrigation (gravity, using hose pipes)\",\n",
      " \"cocoa_irrigation_method_5\": \"Cocoa irrigation method: furrow irrigation (gravity)\",\n",
      " \"cocoa_irrigation_method_6\": \"Cocoa irrigation method: flooding\",\n",
      " \"cocoa_irrigation_method_7\": \"Cocoa irrigation method: other\",\n",
      " \"cocoa_losses_disease_yn\": \"Cocoa was lost to dieseases\",\n",
      " \"cocoa_losses_pests_yn\": \"Cocoa was lost to pests\",\n",
      " \"cocoa_losses_storage_yn\": \"Cocoa was lost during storage\",\n",
      " \"cocoa_losses_disease_type_1\": \"Cocoa disease type: cocoa swollen shoot virus disease (cssvd)\",\n",
      " \"cocoa_losses_disease_type_2\": \"Cocoa disease type: black pod disease\",\n",
      " \"cocoa_losses_disease_type_3\": \"Cocoa disease type: blight thread\",\n",
      " \"cocoa_losses_disease_type_5\": \"Cocoa disease type: le chancre\",\n",
      " \"cocoa_losses_disease_type_4\": \"Cocoa disease type: other\",\n",
      " \"cocoa_losses_disease_affect\": \"Effect of cocoa disease\",\n",
      " \"cocoa_losses_pests_type_1\": \"Cocoa pest type: mistletoe\",\n",
      " \"cocoa_losses_pests_type_2\": \"Cocoa pest type: akate\",\n",
      " \"cocoa_losses_pests_type_3\": \"Cocoa pest type: mirids\",\n",
      " \"cocoa_losses_pests_type_4\": \"Cocoa pest type: black ants\",\n",
      " \"cocoa_losses_pests_type_5\": \"Cocoa pest type: termites\",\n",
      " \"cocoa_losses_pests_type_6\": \"Cocoa pest type: rodents\",\n",
      " \"cocoa_losses_pests_type_7\": \"Cocoa pest type: other\",\n",
      " \"cocoa_losses_pests_type_9\": \"Cocoa pest type: foreuse de tige\",\n",
      " \"cocoa_losses_pests_type_10\": \"Cocoa pest type: mousse verte\",\n",
      " \"cocoa_losses_pests_type_999\": \"Cocoa pest type: don't know\",\n",
      " \"cocoa_losses_pests_affect\": \"Effect of cocoa pest\",\n",
      " \"cocoa_losses_storage_affect\": \"Effect of cocoa loss during storage\",\n",
      " \"cocoa_records_yn\": \"Cocoa records were kept\",\n",
      " \"cocoa_records_type_1\": \"Cocoa records: money spent on inputs\",\n",
      " \"cocoa_records_type_2\": \"Cocoa records: money spent on hired labourers\",\n",
      " \"cocoa_records_type_3\": \"Cocoa records: number of days household members spent labouring\",\n",
      " \"cocoa_records_type_4\": \"Cocoa records: the volume produced\",\n",
      " \"cocoa_records_type_5\": \"Cocoa records: the volume sold, the price and money received\",\n",
      " \"cocoa_records_type_999\": \"Cocoa records: don't know\",\n",
      " \"cocoa_decisions_gender_1\": \"Decision making was done by man\",\n",
      " \"cocoa_decisions_gender_2\": \"Decision making was done by women\",\n",
      " \"cocoa_decisions_gender\": \"Decision making was done by man, women or both\",\n",
      " \"cocoa_sold_share\": \"Cocoa sold (share of total)\",\n",
      " \"cocoa_sold_kg\": \"Cocoa sold (kgs)\",\n",
      " \"cocoa_sold_kgha\": \"Cocoa sold (kgs/ha)\",\n",
      " \"cocoa_consumed_share\": \"Cocoa consumed at home (share of total)\",\n",
      " \"cocoa_consumed_kg\": \"Cocoa consumed at home (kgs)\",\n",
      " \"cocoa_consumed_kgha\": \"Cocoa consumed at home (kgs/ha)\",\n",
      " \"cocoa_oblig_share\": \"Cocoa duty abunu/abusan (share of total)\",\n",
      " \"cocoa_oblig_kg\": \"Cocoa duty abunu/abusan (kgs)\",\n",
      " \"cocoa_oblig_kgha\": \"Cocoa duty abunu/abusan (kgs/ha)\",\n",
      " \"cocoa_givenaway_share\": \"Cocoa given away free (share of total)\",\n",
      " \"cocoa_givenaway_kg\": \"Cocoa given away free (kgs)\",\n",
      " \"cocoa_givenaway_kgha\": \"Cocoa given away free (kgs/ha)\",\n",
      " \"cocoa_barterexchange_share\": \"Cocoa given in exchange for other goods (barter) (share of total)\",\n",
      " \"cocoa_barterexchange_kg\": \"Cocoa given in exchange for other goods (barter) (kgs)\",\n",
      " \"cocoa_barterexchange_kgha\": \"Cocoa given in exchange for other goods (barter) (kgs/ha)\",\n",
      " \"cocoa_givenforlabour_share\": \"Cocoa given as payment for labour (share of total)\",\n",
      " \"cocoa_givenforlabour_kg\": \"Cocoa given as payment for labour (kgs)\",\n",
      " \"cocoa_givenforlabour_kgha\": \"Cocoa given as payment for labour (kgs/ha)\",\n",
      " \"cocoa_lost_share\": \"Cocoa lost (eg from bad quality, rotting, theft) (share of total)\",\n",
      " \"cocoa_lost_kg\": \"Cocoa lost (eg from bad quality, rotting, theft) (kgs)\",\n",
      " \"cocoa_lost_kgha\": \"Cocoa lost (eg from bad quality, rotting, theft) (kgs/ha)\",\n",
      " \"cocoa_price_common_usdkg\": \"Common cocoa price (usd/kg)\",\n",
      " \"cocoa_price_usdkg\": \"Country average cocoa price (usd/kg)\",\n",
      " \"cocoa_price_highest_usdkg\": \"Cocoa highest price (usd/kg)\",\n",
      " \"cocoa_price_lowest_usdkg\": \"Cocoa lowest price (usd/kg)\",\n",
      " \"cocoa_marketing_months_1\": \"Sales of cocoa in: january\",\n",
      " \"cocoa_marketing_months_2\": \"Sales of cocoa in: february\",\n",
      " \"cocoa_marketing_months_3\": \"Sales of cocoa in: march\",\n",
      " \"cocoa_marketing_months_4\": \"Sales of cocoa in: april\",\n",
      " \"cocoa_marketing_months_5\": \"Sales of cocoa in: may\",\n",
      " \"cocoa_marketing_months_6\": \"Sales of cocoa in: june\",\n",
      " \"cocoa_marketing_months_7\": \"Sales of cocoa in: july\",\n",
      " \"cocoa_marketing_months_8\": \"Sales of cocoa in: august\",\n",
      " \"cocoa_marketing_months_9\": \"Sales of cocoa in: september\",\n",
      " \"cocoa_marketing_months_10\": \"Sales of cocoa in: october\",\n",
      " \"cocoa_marketing_months_11\": \"Sales of cocoa in: november\",\n",
      " \"cocoa_marketing_months_12\": \"Sales of cocoa in: december\",\n",
      " \"cocoa_marketing_months_999\": \"Sales of cocoa in: don't know\",\n",
      " \"cocoa_marketing_months_99\": \"Sales of cocoa in: none/ not applicable\",\n",
      " \"cocoa_buyers_1\": \"Cocoa sold to pbc/produce buying company\",\n",
      " \"cocoa_buyers_2\": \"Cocoa sold to aamc/akuafo adamfo\",\n",
      " \"cocoa_buyers_3\": \"Cocoa sold to abl/adwumapa\",\n",
      " \"cocoa_buyers_4\": \"Cocoa sold to olam\",\n",
      " \"cocoa_buyers_5\": \"Cocoa sold to fcl/fedco\",\n",
      " \"cocoa_buyers_6\": \"Cocoa sold to agl/armajaro\",\n",
      " \"cocoa_buyers_7\": \"Cocoa sold to tgl/transroyal\",\n",
      " \"cocoa_buyers_8\": \"Cocoa sold to kkl/kuapa kokoo\",\n",
      " \"cocoa_buyers_9\": \"Cocoa sold to cmgl/cocoa merchants\",\n",
      " \"cocoa_buyers_10\": \"Cocoa sold to rcl/royal commodities\",\n",
      " \"cocoa_buyers_13\": \"Cocoa sold to cemoi\",\n",
      " \"cocoa_buyers_14\": \"Cocoa sold to barry callebault / saco\",\n",
      " \"cocoa_buyers_15\": \"Cocoa sold to cargill\",\n",
      " \"cocoa_buyers_16\": \"Cocoa sold to coop\\uace1tive/groupement/association de producteurs\",\n",
      " \"cocoa_buyers_11\": \"Cocoa sold to other licenced buying company (lbc)\",\n",
      " \"cocoa_buyers_12\": \"Cocoa sold to other non-licenced buyer\",\n",
      " \"cocoa_buyers_17\": \"Cocoa sold to other\",\n",
      " \"cocoa_buyers_999\": \"Cocoa sold to don't know\",\n",
      " \"cocoa_marketing_gender_1\": \"Cocoa was sold by man\",\n",
      " \"cocoa_marketing_gender_2\": \"Cocoa was sold by women\",\n",
      " \"cocoa_marketing_gender\": \"Cocoa was sold by man, women or both\",\n",
      " \"cocoa_marketing_income_yn\": \"Do you know how much income your household earned from cocoa last year?\",\n",
      " \"cocoa_marketing_income_usd\": \"cocoa reported income (usd)\",\n",
      " \"cocoa_marketing_income_usdha\": \"cocoa reported income (usd/ha)\",\n",
      " \"cocoa_training_5years_yn\": \"Received training in the last 5 years\",\n",
      " \"cocoa_training_days_cat\": \"Number of training days related to cocoa in the past 5 years\",\n",
      " \"cocoa_training_lastyear_yn\": \"If training was received in the last 5 years, was training received in the last \",\n",
      " \"cocoa_training_topic_1\": \"Training on good production practices\",\n",
      " \"cocoa_training_topic_2\": \"Training on good environmental practices\",\n",
      " \"cocoa_training_topic_3\": \"Training on pest and disease management\",\n",
      " \"cocoa_training_topic_4\": \"Training on post-harvest storage and processing\",\n",
      " \"cocoa_training_topic_5\": \"Training on health and safety\",\n",
      " \"cocoa_training_topic_6\": \"Training on gender\",\n",
      " \"cocoa_training_topic_7\": \"Training on child labour\",\n",
      " \"cocoa_training_topic_8\": \"Training on record keeping and business management\",\n",
      " \"cocoa_training_topic_9\": \"Training on savings and loans\",\n",
      " \"cocoa_training_topic_10\": \"Training on cooperative management\",\n",
      " \"cocoa_training_topic_11\": \"Training on other\",\n",
      " \"cocoa_training_source_1\": \"Training on farmer organisation\",\n",
      " \"cocoa_training_source_2\": \"Training on government extension\",\n",
      " \"cocoa_training_source_3\": \"Training on trader\",\n",
      " \"cocoa_training_source_4\": \"Training on company, lbc\",\n",
      " \"cocoa_training_source_5\": \"Training on ngo\",\n",
      " \"cocoa_training_source_6\": \"Training on other\",\n",
      " \"cocoa_training_source_999\": \"Training on don't know\",\n",
      " \"cocoa_certification_yn\": \"Certified by utz, ra, fairtrade,  organic or others\",\n",
      " \"cocoa_certification_clean_yn\": \"Certified by utz, ra, fairtrade or organic (removing other values)\",\n",
      " \"cocoa_cert_type_clean_1\": \"Certification: utz\",\n",
      " \"cocoa_cert_type_clean_2\": \"Certification: rainforest alliance\",\n",
      " \"cocoa_cert_type_clean_3\": \"Certification: flo / fairtrade\",\n",
      " \"cocoa_cert_type_clean_4\": \"Certification: organic\",\n",
      " \"cocoa_prodgroup_hhmem\": \"Household is member of cocoa producer group\",\n",
      " \"cocoa_prodgroup_respondent\": \"Respondent is member of cocoa producer group\",\n",
      " \"cocoa_prodgroup_services_1\": \"Producer group services: trainings\",\n",
      " \"cocoa_prodgroup_services_2\": \"Producer group services: providing inputs (eg fertilizer, planting material)\",\n",
      " \"cocoa_prodgroup_services_3\": \"Producer group services: savings facility\",\n",
      " \"cocoa_prodgroup_services_4\": \"Producer group services: loans facility\",\n",
      " \"cocoa_prodgroup_services_5\": \"Producer group services: organising labour\",\n",
      " \"cocoa_prodgroup_services_6\": \"Producer group services: storage of produce\",\n",
      " \"cocoa_prodgroup_services_7\": \"Producer group services: grading of produce\",\n",
      " \"cocoa_prodgroup_services_8\": \"Producer group services: marketing\",\n",
      " \"cocoa_prodgroup_services_9\": \"Producer group services: other\",\n",
      " \"cocoa_prodgroup_services_10\": \"Producer group services: none\",\n",
      " \"cocoa_prodgroup_services_999\": \"Producer group services: don't know\",\n",
      " \"cocoa_credit_yn\": \"Received inputs on credit\",\n",
      " \"cocoa_credit_inputs_1\": \"Inputs on credit: fertilizer\",\n",
      " \"cocoa_credit_inputs_2\": \"Inputs on credit: pesticides, herbicides or fungicides\",\n",
      " \"cocoa_credit_inputs_3\": \"Inputs on credit: seedlings\",\n",
      " \"cocoa_credit_inputs_4\": \"Inputs on credit: other\",\n",
      " \"cocoa_credit_whom\": \"Inputs on credit: source\",\n",
      " \"loan_lastyear_yn\": \"Money borrowed last year\",\n",
      " \"loan_source_1\": \"Loan source: family or friends\",\n",
      " \"loan_source_2\": \"Loan source: village money lender\",\n",
      " \"loan_source_3\": \"Loan source: lbc or pc (licenced buying company or purchasing clerk)\",\n",
      " \"loan_source_4\": \"Loan source: vsla/islc (informal savings and loans group)\",\n",
      " \"loan_source_5\": \"Loan source: savings and credit cooperative (sacco) / credit union\",\n",
      " \"loan_source_6\": \"Loan source: microfinance institution (mfi)\",\n",
      " \"loan_source_7\": \"Loan source: bank\",\n",
      " \"loan_source_8\": \"Loan source: trader\",\n",
      " \"loan_source_9\": \"Loan source: company\",\n",
      " \"loan_source_10\": \"Loan source: church / mosque\",\n",
      " \"loan_source_11\": \"Loan source: other\",\n",
      " \"loan_source_12\": \"Loan source: pisteurs\",\n",
      " \"loan_source_13\": \"Loan source: cooperative, producer's association, group\",\n",
      " \"loan_source_999\": \"Loan source: don't know\",\n",
      " \"loan_amount_usd\": \"Loan Amounts (USD)\",\n",
      " \"loan_repayment_success\": \"Loan repaid\",\n",
      " \"loan_repayment_c\": \"Collateral\",\n",
      " \"loan_use_1\": \"Loan use: seedlings\",\n",
      " \"loan_use_2\": \"Loan use: inputs (fertilizer, herbicides, pesticides)\",\n",
      " \"loan_use_3\": \"Loan use: hired labourers\",\n",
      " \"loan_use_4\": \"Loan use: farm tools\",\n",
      " \"loan_use_5\": \"Loan use: hire machinary, tractor\",\n",
      " \"loan_use_6\": \"Loan use: machinary, tractor\",\n",
      " \"loan_use_7\": \"Loan use: car or motorbike\",\n",
      " \"loan_use_8\": \"Loan use: buy land or lease land\",\n",
      " \"loan_use_9\": \"Loan use: transport costs\",\n",
      " \"loan_use_10\": \"Loan use: storage\",\n",
      " \"loan_use_11\": \"Loan use: repair machinary\",\n",
      " \"loan_use_12\": \"Loan use: processing\",\n",
      " \"loan_use_13\": \"Loan use: farmer organisation membership fees\",\n",
      " \"loan_use_14\": \"Loan use: school fees\",\n",
      " \"loan_use_15\": \"Loan use: clothes\",\n",
      " \"loan_use_16\": \"Loan use: food for consumption\",\n",
      " \"loan_use_17\": \"Loan use: funeral costs\",\n",
      " \"loan_use_19\": \"Loan use: wedding costs\",\n",
      " \"loan_use_20\": \"Loan use: small business / trade\",\n",
      " \"loan_use_18\": \"Loan use: other\",\n",
      " \"loan_use_999\": \"Loan use: don't know\",\n",
      " \"noloan_whynot_1\": \"No loan reason: not available in the community\",\n",
      " \"noloan_whynot_2\": \"No loan reason: interest rate too high\",\n",
      " \"noloan_whynot_3\": \"No loan reason: repay period too short\",\n",
      " \"noloan_whynot_4\": \"No loan reason: lack of collatoral /security\",\n",
      " \"noloan_whynot_5\": \"No loan reason: i didn't need a loan\",\n",
      " \"noloan_whynot_6\": \"No loan reason: i don't want to be in debt\",\n",
      " \"noloan_whynot_7\": \"No loan reason: spouse refused\",\n",
      " \"noloan_whynot_8\": \"No loan reason: other\",\n",
      " \"distance_hometofarm\": \"Distance to your main farm\",\n",
      " \"distance_localmarket\": \"Distance to the nearest local market\",\n",
      " \"distance_inputsupplier\": \"Distance to the nearest input supplier?\",\n",
      " \"distance_healthclinic\": \"Distance to the nearest health clinic\",\n",
      " \"distance_sealed_road\": \"Distance to the nearest tarred road (sealed road)\",\n",
      " \"hh_own_house\": \"House owned\",\n",
      " \"hh_own_land\": \"Land owned\",\n",
      " \"hh_sleep_room\": \"Number of sleeping rooms\",\n",
      " \"hh_members_per_sleep_room\": \"Members per sleeping rooms\",\n",
      " \"hh_drinking_water\": \"Main source of drinking water\",\n",
      " \"hh_toilet\": \"Type of toilet\",\n",
      " \"hh_cooking_fuel\": \"Cooking fuel\",\n",
      " \"hh_roof\": \"Roof material\",\n",
      " \"hh_floor\": \"Floor material\",\n",
      " \"hh_walls\": \"Walls material\",\n",
      " \"hh_bed\": \"Bed\",\n",
      " \"hh_table\": \"Table\",\n",
      " \"hh_cabinet_cupboard\": \"Cabinet/cupboard\",\n",
      " \"hh_sofa\": \"sofa / couch\",\n",
      " \"hh_watch\": \"Watch\",\n",
      " \"hh_wall_clock\": \"Wall clock\",\n",
      " \"hh_sewing_machine\": \"Sewing machine\",\n",
      " \"hh_pots_cookware\": \"Pots and cookware\",\n",
      " \"hh_bicycle\": \"Bicycle\",\n",
      " \"hh_motorcycle_scooter\": \"Motorcycle or Scooter\",\n",
      " \"hh_animal_cart\": \"Animal-drawn cart\",\n",
      " \"hh_tractor\": \"Tracteur\",\n",
      " \"hh_car_truck\": \"Car or Truck\",\n",
      " \"hh_electricity\": \"Electricity\",\n",
      " \"hh_radio\": \"Radio\",\n",
      " \"hh_bw_television\": \"Black & white television\",\n",
      " \"hh_color_television\": \"Color television\",\n",
      " \"hh_television\": \"Television (BW or Color)\",\n",
      " \"hh_cable_subscription\": \"Cable or satelite television\",\n",
      " \"hh_mobile_telephone\": \"Mobile telephone\",\n",
      " \"hh_mobile_telephone_num\": \"Number of mobile phones in the household\",\n",
      " \"hh_telephone\": \"Telephone (non-mobile)\",\n",
      " \"hh_fan\": \"Electric fan\",\n",
      " \"hh_refrigerator\": \"Refrigerator\",\n",
      " \"hh_freezer\": \"Freezer\",\n",
      " \"hh_generator\": \"Generator\",\n",
      " \"hh_washing_machine\": \"Washing machine\",\n",
      " \"hh_gaselectric_stove\": \"Gas or electric stove\",\n",
      " \"hh_iron\": \"Box or electric iron\",\n",
      " \"hh_computer_tablet\": \"Computer/tablet computer\",\n",
      " \"hh_camera\": \"Photo camera ( not on phone )\",\n",
      " \"hh_video_dvd\": \"Video deck/DVD/VCD\",\n",
      " \"hh_internet_access\": \"Internet access\",\n",
      " \"hh_boat_motor\": \"Boat with a motor\",\n",
      " \"hh_boat_no_motor\": \"Boat without a motor\",\n",
      " \"hh_bank_account\": \"Bank account\",\n",
      " \"hh_mosquito_net\": \"Mosquito net\",\n",
      " \"hh_mosquito_net_num\": \"Number of mosquito nets in your household\",\n",
      " \"hh_mosquito_net_lastnight\": \"Did all members of your household sleep under a mosquito net last night?\",\n",
      " \"ls_dairy_cattle\": \"Dairy cattle\",\n",
      " \"ls_cows_bulls\": \"Cows/bulls\",\n",
      " \"ls_horses_donkeys_mules\": \"Horses/donkeys/mules\",\n",
      " \"ls_goats\": \"Goats\",\n",
      " \"ls_pigs\": \"Pigs\",\n",
      " \"ls_rabbits\": \"Rabbits\",\n",
      " \"ls_grasscutter\": \"Grasscutter\",\n",
      " \"ls_sheep\": \"Sheep\",\n",
      " \"ls_chickens\": \"Chickens\",\n",
      " \"ls_other_poultry\": \"Other poultry\",\n",
      " \"ls_other\": \"Other animals\",\n",
      " \"nutrition_a_grains\": \"GRAINS: Maize, wheat or Hausa (spicy millet) porridge; Bread or sometimes buff l\",\n",
      " \"nutrition_b_roots_tubers_plant\": \"ROOTS & TUBERS: Fufu; Banku (fermented corn and cassava dough); Akpele (corn and\",\n",
      " \"nutrition_c_pulses\": \"PULSES: Bean stew; Koose (fried bean cake); Wakye (Rice and beans)\",\n",
      " \"nutrition_d_nuts_seeds\": \"NUTS & SEEDS, groundnut soup, groundnut sauce, peanut, seeds\",\n",
      " \"nutrition_e_milk_products\": \"Milk, cheese, yoghurt or other milk products (but NOT including condensed milk o\",\n",
      " \"nutrition_f_organ_meat\": \"ORGAN MEAT: Liver, kidney, heart or other organ meats or blood-based foods, incl\",\n",
      " \"nutrition_g_meat_poultry\": \"MEAT & POULTRY: Beef, pork, lamb, goat, rabbit, wild game meat, chicken, duck or\",\n",
      " \"nutrition_h_fish_seafood\": \"FISH & SEAFOOD: Fresh or dried fish, shellfish or seafood\",\n",
      " \"nutrition_i_eggs\": \"EGGS: Eggs from poultry or any other bird\",\n",
      " \"nutrition_j_dark_leafy_veg\": \"LEAFY VEGETABLES: Medium-to-dark green leafy vegetables, including wild/foraged \",\n",
      " \"nutrition_k_vita_vegroots\": \"ROOT VEGETABLES: Pumpkin, carrots, squash or sweet potatoes that are yellow or o\",\n",
      " \"nutrition_l_vita_fruits\": \"Red palm oil, ripe mango, ripe papaya;\",\n",
      " \"nutrition_m_other_vegetables\": \"OTHER VEGETABLES: Mainly eggplants, aubergines, tomatoes, okras\",\n",
      " \"nutrition_n_other_fruits\": \"OTHER FRUITS: Mainly oranges and sweet bananas\",\n",
      " \"nutrition_condiments\": \"Ingredients used in small quantities for flavour, such as chilies, spices, herbs\",\n",
      " \"nutrition_other_bevfood\": \"TEA & COFFEE: if not sweetened, clear broth, alcohol, pickles, olives and simila\",\n",
      " \"hh_members_months_away\": \"Number of members how in living away\",\n",
      " \"foodsec1\": \"Meals per day: january\",\n",
      " \"foodsec2\": \"Meals per day: february\",\n",
      " \"foodsec3\": \"Meals per day: march\",\n",
      " \"foodsec4\": \"Meals per day: april\",\n",
      " \"foodsec5\": \"Meals per day: may\",\n",
      " \"foodsec6\": \"Meals per day: june\",\n",
      " \"foodsec7\": \"Meals per day: july\",\n",
      " \"foodsec8\": \"Meals per day: august\",\n",
      " \"foodsec9\": \"Meals per day: september\",\n",
      " \"foodsec10\": \"Meals per day: october\",\n",
      " \"foodsec11\": \"Meals per day: november\",\n",
      " \"foodsec12\": \"Meals per day: december\",\n",
      " \"instanceid\": \"\",\n",
      " \"cocoa_hh_mostimp\": \"Cocoa Farmer (Importance)\",\n",
      " \"migrant\": \"Migrant\",\n",
      " \"MDD_1_W\": \"Group 1: grains, white roots and tubers, and plantains\",\n",
      " \"MDD_2_W\": \"Group 2: pulses (beans, peas and lentils)\",\n",
      " \"MDD_3_W\": \"Group 3: nuts and seeds\",\n",
      " \"MDD_4_W\": \"Group 4: dairy\",\n",
      " \"MDD_5_W\": \"Group 5: meat, poultry and fish\",\n",
      " \"MDD_6_W\": \"Group 6: eggs\",\n",
      " \"MDD_7_W\": \"Group 7: dark green leafy vegetables\",\n",
      " \"MDD_8_W\": \"Group 8: other vitamin a-rich fruits and vegetables\",\n",
      " \"MDD_9_W\": \"Group 9: other vegetables\",\n",
      " \"MDD_10_W\": \"Group 10: other fruits\",\n",
      " \"MDD_score_W\": \"Women's Dietary Diversity Score\",\n",
      " \"MDD_bin_W\": \"Achieved Minimum Dietary Diversity\",\n",
      " \"youth\": \"Youth\",\n",
      " \"hh_income_1\": \"Household income source: sale of cocoa\",\n",
      " \"hh_income_2\": \"Household income source: sale of other crops\",\n",
      " \"hh_income_3\": \"Household income source: sale of livestock or livestock products\",\n",
      " \"hh_income_4\": \"Household income source: sale of fish\",\n",
      " \"hh_income_5\": \"Household income source: sale of bush products (bush meat, charcoal, wood etc)\",\n",
      " \"hh_income_6\": \"Household income source: labouring for other people on their farms\",\n",
      " \"hh_income_7\": \"Household income source: labouring for other people non-agriculture\",\n",
      " \"hh_income_8\": \"Household income source: own small business or trading\",\n",
      " \"hh_income_9\": \"Household income source: salary employment with a company\",\n",
      " \"hh_income_10\": \"Household income source: salary employment in government job (teacher, nurse, po\",\n",
      " \"hh_income_11\": \"Household income source: sale or lease of land\",\n",
      " \"hh_income_12\": \"Household income source: remittances - money given by friends and family who are\",\n",
      " \"hh_income_13\": \"Household income source: other\",\n",
      " \"n_hh_incomes\": \"Number of different household sources of incomes\",\n",
      " \"dhs\": \"DHS Rural Index\",\n",
      " \"dhs_combined\": \"DHS Country Level Score\",\n",
      " \"dhs_quintile\": \"Wealth Quintile, According to DHS Wealth Index\",\n",
      " \"PPI_SCORE\": \"\",\n",
      " \"PPI_100_Nat\": \"Likelihood of being under the 100% National Poverty Line\",\n",
      " \"PPI_125dollar_day\": \"Likelihood of being under $1.25/day PPP 2005\",\n",
      " \"PPI_200dollar_day\": \"Likelihood of being under $2.00/day PPP 2005\",\n",
      " \"PPI_250dollar_day\": \"Likelihood of being under $2.50/day PPP 2005\",\n",
      " \"PPI_190dollar_day\": \"Likelihood of being under $1.90/day PPP 2011\",\n",
      " \"PPI_310dollar_day\": \"Likelihood of being under $3.10/day PPP 2011\",\n",
      " \"non_cocoa_land\": \"Land (ha) used for non-cocoa\",\n",
      " \"share_cocoa_land\": \"Share of cocoa land over all land\",\n",
      " \"foodsec_bin_1\": \"3 meals/day: january\",\n",
      " \"foodsec_bin_2\": \"3 meals/day: february\",\n",
      " \"foodsec_bin_3\": \"3 meals/day: march\",\n",
      " \"foodsec_bin_4\": \"3 meals/day: april\",\n",
      " \"foodsec_bin_5\": \"3 meals/day: may\",\n",
      " \"foodsec_bin_6\": \"3 meals/day: june\",\n",
      " \"foodsec_bin_7\": \"3 meals/day: july\",\n",
      " \"foodsec_bin_8\": \"3 meals/day: august\",\n",
      " \"foodsec_bin_9\": \"3 meals/day: september\",\n",
      " \"foodsec_bin_10\": \"3 meals/day: october\",\n",
      " \"foodsec_bin_11\": \"3 meals/day: november\",\n",
      " \"foodsec_bin_12\": \"3 meals/day: december\"\n",
      "}\n"
     ]
    }
   ],
   "source": [
    "print(json.dumps(data_properties.variable_labels(), indent=1))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The full dataset has (too) many features (1252), in comparison to the number of observations. \n",
    "\n",
    "This includes features which are not useful for the analysis, and some are sets of dummy variables (which is fine).\n",
    "\n",
    "I will initially drop features that I know I will not be needing for any of the analysis.\n",
    "\n",
    "They include:\n",
    "- location variables, except country. Analysis will be on a country level\n",
    "- consent indicator, as this should be 'yes' for all observations included in this dataset\n",
    "- instanceid, unique identifier\n",
    "- data on the respondent (data on the head of the household and on the full household will suffice)\n",
    "- variables in imputed values (\"fm_\"), since the imputation was done based on the country level, and I will do this by other \n",
    "subgroups\n",
    "- variables with the profit and value of production, since these will be re-calculated here\n",
    "- details on the price, since only the common price per kg in USD will be used\n",
    "- indicator of selling crops for the subset of those that produced: this variable carries by construction many missing values (those that did not produce) and therefore is not of interest\n",
    "- variables with daily rates for hired labour, since the original reports refer they used median hiring cost to adjust for noise in the responses\n",
    "- question on the \"(days)known\" for labour activities or other detailed, since these are intermediary variables to filter on the amount of days and therefore not needed for this analysis\n",
    "- land unit prefence, since all is already converted into ha\n",
    "- cocoa_certification_yn, since there is already a cocoa_certification_clean_yn which is a better version\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Confirm `consent` is `yes` for all observations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "yes    3045\n",
       "Name: consent_yn, dtype: int64"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['consent_yn'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Drop variables as explained above:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(['region_comparable', 'up_region', 'sub_region', 'village', 'latitude', 'longitude'], axis = 1, inplace = True) \n",
    "data.drop(['today','consent_yn', 'instanceid', 'land_unit_preference'], axis = 1, inplace = True)\n",
    "data.drop(list(data.filter(regex = 'p1_')), axis = 1, inplace = True) \n",
    "data.drop(['age_cat', 'leader','cocoa_prodgroup_respondent'], axis = 1, inplace = True) \n",
    "data.drop(list(data.filter(regex = 'fm_')), axis = 1, inplace = True) \n",
    "data.drop(['cocoa_price_common_usdkg','cocoa_price_highest_usdkg','cocoa_price_lowest_usdkg'], axis = 1, inplace = True) \n",
    "data.drop(['cocoa_valueproduction_usd','cocoa_valueproduction_usd_ha','cocoa_profit1_usdha','cocoa_profit2_usdha'], axis = 1, inplace = True) \n",
    "data.drop(list(data.filter(regex = 'crops_sold_ifprod')), axis = 1, inplace = True) \n",
    "data.drop(list(data.filter(regex = 'orig_hi_cost')), axis = 1, inplace = True)\n",
    "data.drop(list(data.filter(regex = 'cocoa_marketing_income_yn')), axis = 1, inplace = True) \n",
    "data.drop(list(data.filter(regex = 'days_known_yn')), axis = 1, inplace = True) \n",
    "data.drop(['cocoa_certification_yn'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For each production activity there is also two sex of workers indicators (`_gender_1, _gender_2`), plus a combined indicator (`_gender`). I can drop the later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['head_gender',\n",
       " 'cocoa_landclear_gender_1',\n",
       " 'cocoa_landclear_gender_2',\n",
       " 'cocoa_landclear_gender',\n",
       " 'cocoa_landprep_gender_1',\n",
       " 'cocoa_landprep_gender_2',\n",
       " 'cocoa_landprep_gender',\n",
       " 'cocoa_planting_gender_1',\n",
       " 'cocoa_planting_gender_2',\n",
       " 'cocoa_planting_gender',\n",
       " 'cocoa_gfert_gender_1',\n",
       " 'cocoa_gfert_gender_2',\n",
       " 'cocoa_gfert_gender',\n",
       " 'cocoa_lfert_gender_1',\n",
       " 'cocoa_lfert_gender_2',\n",
       " 'cocoa_lfert_gender',\n",
       " 'cocoa_manurecomp_gender_1',\n",
       " 'cocoa_manurecomp_gender_2',\n",
       " 'cocoa_manurecomp_gender',\n",
       " 'cocoa_herb_gender_1',\n",
       " 'cocoa_herb_gender_2',\n",
       " 'cocoa_herb_gender',\n",
       " 'cocoa_pest_gender_1',\n",
       " 'cocoa_pest_gender_2',\n",
       " 'cocoa_pest_gender',\n",
       " 'cocoa_fung_gender_1',\n",
       " 'cocoa_fung_gender_2',\n",
       " 'cocoa_fung_gender',\n",
       " 'cocoa_weed_gender_1',\n",
       " 'cocoa_weed_gender_2',\n",
       " 'cocoa_weed_gender',\n",
       " 'cocoa_pruning_gender_1',\n",
       " 'cocoa_pruning_gender_2',\n",
       " 'cocoa_pruning_gender',\n",
       " 'cocoa_harvest_gender_1',\n",
       " 'cocoa_harvest_gender_2',\n",
       " 'cocoa_harvest_gender',\n",
       " 'cocoa_podbreak_gender_1',\n",
       " 'cocoa_podbreak_gender_2',\n",
       " 'cocoa_podbreak_gender',\n",
       " 'cocoa_ferment_gender_1',\n",
       " 'cocoa_ferment_gender_2',\n",
       " 'cocoa_ferment_gender',\n",
       " 'cocoa_transport_gender_1',\n",
       " 'cocoa_transport_gender_2',\n",
       " 'cocoa_transport_gender',\n",
       " 'cocoa_dry_gender_1',\n",
       " 'cocoa_dry_gender_2',\n",
       " 'cocoa_decisions_gender_1',\n",
       " 'cocoa_decisions_gender_2',\n",
       " 'cocoa_decisions_gender',\n",
       " 'cocoa_marketing_gender_1',\n",
       " 'cocoa_marketing_gender_2',\n",
       " 'cocoa_marketing_gender']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(data.filter(regex = 'gender'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop([\n",
    " 'cocoa_landclear_gender',\n",
    " 'cocoa_landprep_gender',\n",
    " 'cocoa_planting_gender',\n",
    " 'cocoa_gfert_gender',\n",
    " 'cocoa_lfert_gender',\n",
    " 'cocoa_manurecomp_gender',\n",
    " 'cocoa_herb_gender',\n",
    " 'cocoa_pest_gender',\n",
    " 'cocoa_fung_gender',\n",
    " 'cocoa_weed_gender',\n",
    " 'cocoa_pruning_gender',\n",
    " 'cocoa_harvest_gender',\n",
    " 'cocoa_podbreak_gender',\n",
    " 'cocoa_ferment_gender',\n",
    " 'cocoa_transport_gender',\n",
    " 'cocoa_decisions_gender',\n",
    " 'cocoa_marketing_gender'\n",
    "], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly, for each production activity there is also four types of workers indicators (`hhhiredcom_`), plus a combined indicator (`hh_hi`). I can drop the latter."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(list(data.filter(regex = 'hh_hi')), axis = 1, inplace = True) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are 3 sets of features about the allocation of production: in kg, in kg/ha and as share of total. I'll keep only the one on shares."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop([\n",
    " 'cocoa_sold_kg',\n",
    " 'cocoa_sold_kgha',\n",
    " 'cocoa_consumed_kg',\n",
    " 'cocoa_consumed_kgha',\n",
    " 'cocoa_oblig_kg',\n",
    " 'cocoa_oblig_kgha',\n",
    " 'cocoa_givenaway_kg',\n",
    " 'cocoa_givenaway_kgha',\n",
    " 'cocoa_barterexchange_kg',\n",
    " 'cocoa_barterexchange_kgha',\n",
    " 'cocoa_givenforlabour_kg',\n",
    " 'cocoa_givenforlabour_kgha',\n",
    " 'cocoa_lost_kg',\n",
    " 'cocoa_lost_kgha'\n",
    "], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the analysis, I will only look at cocoa households that know their land and production"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data[data['cocoa_hh']=='yes']\n",
    "data = data[data['cocoa_land_known_yn']=='yes']\n",
    "data = data[data['cocoa_prod_known_yn']=='yes']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data[data['cocoa_prod_total_kgsha'].notnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(['cocoa_hh', 'cocoa_land_known_yn','cocoa_prod_known_yn'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Further, for this analysis, I'm only interested in the total production over an year, and not disaggregated into seasons"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop([\n",
    "    'cocoa_season1_start',\n",
    "    'cocoa_season1_end',\n",
    "    'cocoa_season2_start',\n",
    "    'cocoa_season2_end',\n",
    "    'cocoa_prod_season1_kg',\n",
    "    'cocoa_prod_season1_kgha',\n",
    "    'cocoa_prod_season2_kg',\n",
    "    'cocoa_prod_season2_kgha'\n",
    "], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now I can also drop some of the other household type indicators, since I am only looking at cocoa households"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['coconut_hh',\n",
       " 'cocoyam_hh',\n",
       " 'coffee_hh',\n",
       " 'courgette_hh',\n",
       " 'cotton_hh',\n",
       " 'cow_pea_hh',\n",
       " 'eggplant_hh',\n",
       " 'fodder_hh',\n",
       " 'garlic_hh',\n",
       " 'ginger_hh',\n",
       " 'groundnuts_hh',\n",
       " 'lemon_hh',\n",
       " 'lime_hh',\n",
       " 'maize_hh',\n",
       " 'mango_hh',\n",
       " 'millet_hh',\n",
       " 'okra_hh',\n",
       " 'onions_hh',\n",
       " 'oranges_hh',\n",
       " 'palm_hh',\n",
       " 'papaya_hh',\n",
       " 'peas_hh',\n",
       " 'peppers_hh',\n",
       " 'pineapple_hh',\n",
       " 'plantai_hh',\n",
       " 'irish_potato_hh',\n",
       " 'sweet_potato_hh',\n",
       " 'rice_hh',\n",
       " 'rubber_hh',\n",
       " 'spinach_hh',\n",
       " 'sorghum_hh',\n",
       " 'shea_nut_hh',\n",
       " 'sugar_cane_hh',\n",
       " 'tangerine_hh',\n",
       " 'tomatoes_hh',\n",
       " 'watermelon_hh',\n",
       " 'yam_hh',\n",
       " 'avocado_hh_5ago',\n",
       " 'bananas_hh_5ago',\n",
       " 'soya_beans_hh_5ago',\n",
       " 'beans_hh_5ago',\n",
       " 'cabbages_hh_5ago',\n",
       " 'carrots_hh_5ago',\n",
       " 'cashews_hh_5ago',\n",
       " 'cassava_hh_5ago',\n",
       " 'chili_hh_5ago',\n",
       " 'cocoa_hh_5ago',\n",
       " 'coconut_hh_5ago',\n",
       " 'cocoyam_hh_5ago',\n",
       " 'coffee_hh_5ago',\n",
       " 'courgette_hh_5ago',\n",
       " 'cotton_hh_5ago',\n",
       " 'cow_pea_hh_5ago',\n",
       " 'eggplant_hh_5ago',\n",
       " 'fodder_hh_5ago',\n",
       " 'garlic_hh_5ago',\n",
       " 'ginger_hh_5ago',\n",
       " 'groundnuts_hh_5ago',\n",
       " 'lemon_hh_5ago',\n",
       " 'lime_hh_5ago',\n",
       " 'maize_hh_5ago',\n",
       " 'mango_hh_5ago',\n",
       " 'millet_hh_5ago',\n",
       " 'okra_hh_5ago',\n",
       " 'onions_hh_5ago',\n",
       " 'oranges_hh_5ago',\n",
       " 'palm_hh_5ago',\n",
       " 'papaya_hh_5ago',\n",
       " 'peas_hh_5ago',\n",
       " 'peppers_hh_5ago',\n",
       " 'pineapple_hh_5ago',\n",
       " 'plantai_hh_5ago',\n",
       " 'irish_potato_hh_5ago',\n",
       " 'sweet_potato_hh_5ago',\n",
       " 'rice_hh_5ago',\n",
       " 'rubber_hh_5ago',\n",
       " 'spinach_hh_5ago',\n",
       " 'sorghum_hh_5ago',\n",
       " 'shea_nut_hh_5ago',\n",
       " 'sugar_cane_hh_5ago',\n",
       " 'tangerine_hh_5ago',\n",
       " 'tomatoes_hh_5ago',\n",
       " 'watermelon_hh_5ago',\n",
       " 'yam_hh_5ago',\n",
       " 'avocado_hh_change',\n",
       " 'bananas_hh_change',\n",
       " 'soya_beans_hh_change',\n",
       " 'beans_hh_change',\n",
       " 'cabbages_hh_change',\n",
       " 'carrots_hh_change',\n",
       " 'cashews_hh_change',\n",
       " 'cassava_hh_change',\n",
       " 'chili_hh_change',\n",
       " 'cocoa_hh_change',\n",
       " 'coconut_hh_change',\n",
       " 'cocoyam_hh_change',\n",
       " 'coffee_hh_change',\n",
       " 'courgette_hh_change',\n",
       " 'cotton_hh_change',\n",
       " 'cow_pea_hh_change',\n",
       " 'eggplant_hh_change',\n",
       " 'fodder_hh_change',\n",
       " 'garlic_hh_change',\n",
       " 'ginger_hh_change',\n",
       " 'groundnuts_hh_change',\n",
       " 'lemon_hh_change',\n",
       " 'lime_hh_change',\n",
       " 'maize_hh_change',\n",
       " 'mango_hh_change',\n",
       " 'millet_hh_change',\n",
       " 'okra_hh_change',\n",
       " 'onions_hh_change',\n",
       " 'oranges_hh_change',\n",
       " 'palm_hh_change',\n",
       " 'papaya_hh_change',\n",
       " 'peas_hh_change',\n",
       " 'peppers_hh_change',\n",
       " 'pineapple_hh_change',\n",
       " 'plantai_hh_change',\n",
       " 'irish_potato_hh_change',\n",
       " 'sweet_potato_hh_change',\n",
       " 'rice_hh_change',\n",
       " 'rubber_hh_change',\n",
       " 'spinach_hh_change',\n",
       " 'sorghum_hh_change',\n",
       " 'shea_nut_hh_change',\n",
       " 'sugar_cane_hh_change',\n",
       " 'tangerine_hh_change',\n",
       " 'tomatoes_hh_change',\n",
       " 'watermelon_hh_change',\n",
       " 'yam_hh_change',\n",
       " 'cocoa_landclear_hh_n_ha',\n",
       " 'cocoa_landprep_hh_n_ha',\n",
       " 'cocoa_planting_hh_n_ha',\n",
       " 'cocoa_gfert_hh_n_ha',\n",
       " 'cocoa_lfert_hh_n_ha',\n",
       " 'cocoa_manurecomp_hh_n_ha',\n",
       " 'cocoa_herb_hh_n_ha',\n",
       " 'cocoa_pest_hh_n_ha',\n",
       " 'cocoa_fung_hh_n_ha',\n",
       " 'cocoa_weed_hh_n_ha',\n",
       " 'cocoa_pruning_hh_n_ha',\n",
       " 'cocoa_harvest_hh_n_ha',\n",
       " 'cocoa_podbreak_hh_n_ha',\n",
       " 'cocoa_transport_hh_n_ha',\n",
       " 'cocoa_landclear_hhhiredcom_1',\n",
       " 'cocoa_landclear_hhhiredcom_2',\n",
       " 'cocoa_landclear_hhhiredcom_3',\n",
       " 'cocoa_landclear_hhhiredcom_4',\n",
       " 'cocoa_landprep_hhhiredcom_1',\n",
       " 'cocoa_landprep_hhhiredcom_2',\n",
       " 'cocoa_landprep_hhhiredcom_3',\n",
       " 'cocoa_landprep_hhhiredcom_4',\n",
       " 'cocoa_planting_hhhiredcom_1',\n",
       " 'cocoa_planting_hhhiredcom_2',\n",
       " 'cocoa_planting_hhhiredcom_3',\n",
       " 'cocoa_planting_hhhiredcom_4',\n",
       " 'cocoa_gfert_hhhiredcom_1',\n",
       " 'cocoa_gfert_hhhiredcom_2',\n",
       " 'cocoa_gfert_hhhiredcom_3',\n",
       " 'cocoa_gfert_hhhiredcom_4',\n",
       " 'cocoa_lfert_hhhiredcom_1',\n",
       " 'cocoa_lfert_hhhiredcom_2',\n",
       " 'cocoa_lfert_hhhiredcom_3',\n",
       " 'cocoa_lfert_hhhiredcom_4',\n",
       " 'cocoa_manurecomp_hhhiredcom_1',\n",
       " 'cocoa_manurecomp_hhhiredcom_2',\n",
       " 'cocoa_manurecomp_hhhiredcom_3',\n",
       " 'cocoa_manurecomp_hhhiredcom_4',\n",
       " 'cocoa_herb_hhhiredcom_1',\n",
       " 'cocoa_herb_hhhiredcom_2',\n",
       " 'cocoa_herb_hhhiredcom_3',\n",
       " 'cocoa_herb_hhhiredcom_4',\n",
       " 'cocoa_pest_hhhiredcom_1',\n",
       " 'cocoa_pest_hhhiredcom_2',\n",
       " 'cocoa_pest_hhhiredcom_3',\n",
       " 'cocoa_pest_hhhiredcom_4',\n",
       " 'cocoa_fung_hhhiredcom_1',\n",
       " 'cocoa_fung_hhhiredcom_2',\n",
       " 'cocoa_fung_hhhiredcom_3',\n",
       " 'cocoa_fung_hhhiredcom_4',\n",
       " 'cocoa_weed_hhhiredcom_1',\n",
       " 'cocoa_weed_hhhiredcom_2',\n",
       " 'cocoa_weed_hhhiredcom_3',\n",
       " 'cocoa_weed_hhhiredcom_4',\n",
       " 'cocoa_pruning_hhhiredcom_1',\n",
       " 'cocoa_pruning_hhhiredcom_2',\n",
       " 'cocoa_pruning_hhhiredcom_3',\n",
       " 'cocoa_pruning_hhhiredcom_4',\n",
       " 'cocoa_harvest_hhhiredcom_1',\n",
       " 'cocoa_harvest_hhhiredcom_2',\n",
       " 'cocoa_harvest_hhhiredcom_3',\n",
       " 'cocoa_harvest_hhhiredcom_4',\n",
       " 'cocoa_podbreak_hhhiredcom_1',\n",
       " 'cocoa_podbreak_hhhiredcom_2',\n",
       " 'cocoa_podbreak_hhhiredcom_3',\n",
       " 'cocoa_podbreak_hhhiredcom_4',\n",
       " 'cocoa_ferment_hhhiredcom_1',\n",
       " 'cocoa_ferment_hhhiredcom_2',\n",
       " 'cocoa_ferment_hhhiredcom_3',\n",
       " 'cocoa_ferment_hhhiredcom_4',\n",
       " 'cocoa_transport_hhhiredcom_1',\n",
       " 'cocoa_transport_hhhiredcom_2',\n",
       " 'cocoa_transport_hhhiredcom_3',\n",
       " 'cocoa_transport_hhhiredcom_4',\n",
       " 'cocoa_dry_hhhiredcom_1',\n",
       " 'cocoa_dry_hhhiredcom_2',\n",
       " 'cocoa_dry_hhhiredcom_3',\n",
       " 'cocoa_dry_hhhiredcom_4',\n",
       " 'cocoa_prodgroup_hhmem',\n",
       " 'cocoa_hh_mostimp',\n",
       " 'n_hh_incomes']"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(data.filter(regex = '_hh'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(\n",
    "['coconut_hh',\n",
    " 'cocoyam_hh',\n",
    " 'coffee_hh',\n",
    " 'courgette_hh',\n",
    " 'cotton_hh',\n",
    " 'cow_pea_hh',\n",
    " 'eggplant_hh',\n",
    " 'fodder_hh',\n",
    " 'garlic_hh',\n",
    " 'ginger_hh',\n",
    " 'groundnuts_hh',\n",
    " 'lemon_hh',\n",
    " 'lime_hh',\n",
    " 'maize_hh',\n",
    " 'mango_hh',\n",
    " 'millet_hh',\n",
    " 'okra_hh',\n",
    " 'onions_hh',\n",
    " 'oranges_hh',\n",
    " 'palm_hh',\n",
    " 'papaya_hh',\n",
    " 'peas_hh',\n",
    " 'peppers_hh',\n",
    " 'pineapple_hh',\n",
    " 'plantai_hh',\n",
    " 'irish_potato_hh',\n",
    " 'sweet_potato_hh',\n",
    " 'rice_hh',\n",
    " 'rubber_hh',\n",
    " 'spinach_hh',\n",
    " 'sorghum_hh',\n",
    " 'shea_nut_hh',\n",
    " 'sugar_cane_hh',\n",
    " 'tangerine_hh',\n",
    " 'tomatoes_hh',\n",
    " 'watermelon_hh',\n",
    " 'yam_hh',\n",
    " 'avocado_hh_5ago',\n",
    " 'bananas_hh_5ago',\n",
    " 'soya_beans_hh_5ago',\n",
    " 'beans_hh_5ago',\n",
    " 'cabbages_hh_5ago',\n",
    " 'carrots_hh_5ago',\n",
    " 'cashews_hh_5ago',\n",
    " 'cassava_hh_5ago',\n",
    " 'chili_hh_5ago',\n",
    " 'cocoa_hh_5ago',\n",
    " 'coconut_hh_5ago',\n",
    " 'cocoyam_hh_5ago',\n",
    " 'coffee_hh_5ago',\n",
    " 'courgette_hh_5ago',\n",
    " 'cotton_hh_5ago',\n",
    " 'cow_pea_hh_5ago',\n",
    " 'eggplant_hh_5ago',\n",
    " 'fodder_hh_5ago',\n",
    " 'garlic_hh_5ago',\n",
    " 'ginger_hh_5ago',\n",
    " 'groundnuts_hh_5ago',\n",
    " 'lemon_hh_5ago',\n",
    " 'lime_hh_5ago',\n",
    " 'maize_hh_5ago',\n",
    " 'mango_hh_5ago',\n",
    " 'millet_hh_5ago',\n",
    " 'okra_hh_5ago',\n",
    " 'onions_hh_5ago',\n",
    " 'oranges_hh_5ago',\n",
    " 'palm_hh_5ago',\n",
    " 'papaya_hh_5ago',\n",
    " 'peas_hh_5ago',\n",
    " 'peppers_hh_5ago',\n",
    " 'pineapple_hh_5ago',\n",
    " 'plantai_hh_5ago',\n",
    " 'irish_potato_hh_5ago',\n",
    " 'sweet_potato_hh_5ago',\n",
    " 'rice_hh_5ago',\n",
    " 'rubber_hh_5ago',\n",
    " 'spinach_hh_5ago',\n",
    " 'sorghum_hh_5ago',\n",
    " 'shea_nut_hh_5ago',\n",
    " 'sugar_cane_hh_5ago',\n",
    " 'tangerine_hh_5ago',\n",
    " 'tomatoes_hh_5ago',\n",
    " 'watermelon_hh_5ago',\n",
    " 'yam_hh_5ago',\n",
    " 'avocado_hh_change',\n",
    " 'bananas_hh_change',\n",
    " 'soya_beans_hh_change',\n",
    " 'beans_hh_change',\n",
    " 'cabbages_hh_change',\n",
    " 'carrots_hh_change',\n",
    " 'cashews_hh_change',\n",
    " 'cassava_hh_change',\n",
    " 'chili_hh_change',\n",
    " 'cocoa_hh_change',\n",
    " 'coconut_hh_change',\n",
    " 'cocoyam_hh_change',\n",
    " 'coffee_hh_change',\n",
    " 'courgette_hh_change',\n",
    " 'cotton_hh_change',\n",
    " 'cow_pea_hh_change',\n",
    " 'eggplant_hh_change',\n",
    " 'fodder_hh_change',\n",
    " 'garlic_hh_change',\n",
    " 'ginger_hh_change',\n",
    " 'groundnuts_hh_change',\n",
    " 'lemon_hh_change',\n",
    " 'lime_hh_change',\n",
    " 'maize_hh_change',\n",
    " 'mango_hh_change',\n",
    " 'millet_hh_change',\n",
    " 'okra_hh_change',\n",
    " 'onions_hh_change',\n",
    " 'oranges_hh_change',\n",
    " 'palm_hh_change',\n",
    " 'papaya_hh_change',\n",
    " 'peas_hh_change',\n",
    " 'peppers_hh_change',\n",
    " 'pineapple_hh_change',\n",
    " 'plantai_hh_change',\n",
    " 'irish_potato_hh_change',\n",
    " 'sweet_potato_hh_change',\n",
    " 'rice_hh_change',\n",
    " 'rubber_hh_change',\n",
    " 'spinach_hh_change',\n",
    " 'sorghum_hh_change',\n",
    " 'shea_nut_hh_change',\n",
    " 'sugar_cane_hh_change',\n",
    " 'tangerine_hh_change',\n",
    " 'tomatoes_hh_change',\n",
    " 'watermelon_hh_change',\n",
    " 'yam_hh_change']\n",
    ", axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Shape of data set: (1455, 871)\n"
     ]
    }
   ],
   "source": [
    "print(\"Shape of data set:\", data.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check the distribution across countries:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ghana    1008\n",
       "CdI       447\n",
       "Name: country, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['country'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 0.3 Analytical Groups"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In line with the Gap Analysis studies ([here](https://c69aa8ac-6965-42b2-abb7-0f0b86c23d2e.filesusr.com/ugd/0c5ab3_fc3386a550b94a898c7757ee13ab59e6.pdf) and [here](https://c69aa8ac-6965-42b2-abb7-0f0b86c23d2e.filesusr.com/ugd/0c5ab3_93560a9b816d40c3a28daaa686e972a5.pdf)), I will create 3 groups per country:\n",
    "\n",
    "* Female-headed\n",
    "* Male-headed, typical: Productive cocoa land up to 4 ha\n",
    "* Male-headed, large: Productive cocoa land larger than  4ha"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[(data['head_gender'] == 'female') & (data['country'] == 'Ghana') , 'Analytical Group'] = 'Ghana: Female-headed'\n",
    "data.loc[(data['head_gender'] == 'male'  ) & (data['country'] == 'Ghana') & (data['cocoa_land_used_morethan5_ha']<=4), 'Analytical Group'] = 'Ghana: Male-headed, typical'\n",
    "data.loc[(data['head_gender'] == 'male'  ) & (data['country'] == 'Ghana') & (data['cocoa_land_used_morethan5_ha']>4 ), 'Analytical Group'] = 'Ghana: Male-headed, large'\n",
    "\n",
    "data.loc[(data['head_gender'] == 'female') & (data['country'] == 'CdI' ) , 'Analytical Group'] = 'CdI: Female-headed'\n",
    "data.loc[(data['head_gender'] == 'male'  ) & (data['country'] == 'CdI' ) & (data['cocoa_land_used_morethan5_ha']<=4), 'Analytical Group'] = 'CdI: Male-headed, typical'\n",
    "data.loc[(data['head_gender'] == 'male'  ) & (data['country'] == 'CdI' ) & (data['cocoa_land_used_morethan5_ha']>4 ), 'Analytical Group'] = 'CdI: Male-headed, large'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ghana: Male-headed, typical    627\n",
       "CdI: Male-headed, typical      313\n",
       "Ghana: Male-headed, large      219\n",
       "Ghana: Female-headed           162\n",
       "CdI: Male-headed, large        119\n",
       "CdI: Female-headed              15\n",
       "Name: Analytical Group, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Analytical Group'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Because the group `CdI: Female-headed` is too small ( _N = 15_ ) I will remove it from the analysis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data[data['Analytical Group']!='CdI: Female-headed']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 1. Calculating the total income"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.1 Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We calculate the total household income as:\n",
    "\n",
    "_Total Household Income = (net cocoa income)/(income share from cocoa)_\n",
    "\n",
    "In other words, I calculate the net cocoa income, and extrapolate the household income by using the information on the share of total income coming from cocoa\n",
    "\n",
    "Te _net cocoa income_ is calculated as:\n",
    "\n",
    "_net cocoa_income = (total production (kg/ha) * price (USD/kg) - input costs (USD/ha) - labor costs (USD/ha))* Productive land (ha)_\n",
    "\n",
    "To calculate input costs and labor costs I need to go over the different input types and production activities"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.2 Calculating income components"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The detailed cocoa production features are built over a few production activities, types of labor and inputs. I will create lists with those activities, types of labor and inputs for further iteration in this notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "production_activities = [\n",
    "    'landclear',\n",
    "    'landprep',\n",
    "    'planting',\n",
    "    'gfert',\n",
    "    'lfert',\n",
    "    'manurecomp',\n",
    "    'herb',\n",
    "    'pest',\n",
    "    'fung',\n",
    "    'weed',\n",
    "    'pruning',\n",
    "    'harvest',\n",
    "    'podbreak',\n",
    "    'ferment',\n",
    "    'transport',\n",
    "    'dry'\n",
    "]\n",
    "\n",
    "types_of_labor = [\n",
    "    'hi_n_ha',\n",
    "    'hh_n_ha',\n",
    "    'com_n_ha',\n",
    "    'lab_n_ha',\n",
    "    'hi_lab_usdha'\n",
    "]\n",
    "\n",
    "inputs = [\n",
    "    'gfert',\n",
    "    'lfert',\n",
    "    'herb',\n",
    "    'pest',\n",
    "    'fung'\n",
    "]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's first check if these variables have missing values. I look at the `hi_lab_usdha` features, since this is directly relevant for the calculations:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>cocoa_landclear_hi_lab_usdha</th>\n",
       "      <th>cocoa_landprep_hi_lab_usdha</th>\n",
       "      <th>cocoa_planting_hi_lab_usdha</th>\n",
       "      <th>cocoa_gfert_hi_lab_usdha</th>\n",
       "      <th>cocoa_lfert_hi_lab_usdha</th>\n",
       "      <th>cocoa_manurecomp_hi_lab_usdha</th>\n",
       "      <th>cocoa_herb_hi_lab_usdha</th>\n",
       "      <th>cocoa_pest_hi_lab_usdha</th>\n",
       "      <th>cocoa_fung_hi_lab_usdha</th>\n",
       "      <th>cocoa_weed_hi_lab_usdha</th>\n",
       "      <th>cocoa_pruning_hi_lab_usdha</th>\n",
       "      <th>cocoa_harvest_hi_lab_usdha</th>\n",
       "      <th>cocoa_podbreak_hi_lab_usdha</th>\n",
       "      <th>cocoa_transport_hi_lab_usdha</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1214.000000</td>\n",
       "      <td>1177.000000</td>\n",
       "      <td>848.000000</td>\n",
       "      <td>1063.000000</td>\n",
       "      <td>1150.000000</td>\n",
       "      <td>858.000000</td>\n",
       "      <td>1041.000000</td>\n",
       "      <td>969.000000</td>\n",
       "      <td>963.000000</td>\n",
       "      <td>732.000000</td>\n",
       "      <td>598.000000</td>\n",
       "      <td>662.000000</td>\n",
       "      <td>641.000000</td>\n",
       "      <td>932.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>4.767475</td>\n",
       "      <td>1.937751</td>\n",
       "      <td>1.512467</td>\n",
       "      <td>1.013818</td>\n",
       "      <td>4.178011</td>\n",
       "      <td>0.297832</td>\n",
       "      <td>3.168064</td>\n",
       "      <td>9.601581</td>\n",
       "      <td>3.686724</td>\n",
       "      <td>35.093170</td>\n",
       "      <td>7.842022</td>\n",
       "      <td>13.991180</td>\n",
       "      <td>1.790963</td>\n",
       "      <td>4.574852</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>14.252471</td>\n",
       "      <td>10.797291</td>\n",
       "      <td>7.526702</td>\n",
       "      <td>4.708241</td>\n",
       "      <td>12.450053</td>\n",
       "      <td>2.725615</td>\n",
       "      <td>13.565760</td>\n",
       "      <td>17.745373</td>\n",
       "      <td>11.614023</td>\n",
       "      <td>54.913971</td>\n",
       "      <td>18.787256</td>\n",
       "      <td>30.327866</td>\n",
       "      <td>8.157135</td>\n",
       "      <td>12.273155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.980000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>11.620000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>50.590759</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>15.488146</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.988000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>129.067871</td>\n",
       "      <td>160.617813</td>\n",
       "      <td>90.347519</td>\n",
       "      <td>51.627155</td>\n",
       "      <td>99.382263</td>\n",
       "      <td>38.720364</td>\n",
       "      <td>142.297333</td>\n",
       "      <td>116.161095</td>\n",
       "      <td>103.254311</td>\n",
       "      <td>322.669739</td>\n",
       "      <td>125.406960</td>\n",
       "      <td>206.508606</td>\n",
       "      <td>82.280777</td>\n",
       "      <td>103.254303</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       cocoa_landclear_hi_lab_usdha  cocoa_landprep_hi_lab_usdha  \\\n",
       "count                   1214.000000                  1177.000000   \n",
       "mean                       4.767475                     1.937751   \n",
       "std                       14.252471                    10.797291   \n",
       "min                        0.000000                     0.000000   \n",
       "25%                        0.000000                     0.000000   \n",
       "50%                        0.000000                     0.000000   \n",
       "75%                        0.000000                     0.000000   \n",
       "max                      129.067871                   160.617813   \n",
       "\n",
       "       cocoa_planting_hi_lab_usdha  cocoa_gfert_hi_lab_usdha  \\\n",
       "count                   848.000000               1063.000000   \n",
       "mean                      1.512467                  1.013818   \n",
       "std                       7.526702                  4.708241   \n",
       "min                       0.000000                  0.000000   \n",
       "25%                       0.000000                  0.000000   \n",
       "50%                       0.000000                  0.000000   \n",
       "75%                       0.000000                  0.000000   \n",
       "max                      90.347519                 51.627155   \n",
       "\n",
       "       cocoa_lfert_hi_lab_usdha  cocoa_manurecomp_hi_lab_usdha  \\\n",
       "count               1150.000000                     858.000000   \n",
       "mean                   4.178011                       0.297832   \n",
       "std                   12.450053                       2.725615   \n",
       "min                    0.000000                       0.000000   \n",
       "25%                    0.000000                       0.000000   \n",
       "50%                    0.000000                       0.000000   \n",
       "75%                    0.000000                       0.000000   \n",
       "max                   99.382263                      38.720364   \n",
       "\n",
       "       cocoa_herb_hi_lab_usdha  cocoa_pest_hi_lab_usdha  \\\n",
       "count              1041.000000               969.000000   \n",
       "mean                  3.168064                 9.601581   \n",
       "std                  13.565760                17.745373   \n",
       "min                   0.000000                 0.000000   \n",
       "25%                   0.000000                 0.000000   \n",
       "50%                   0.000000                 0.000000   \n",
       "75%                   0.000000                11.620000   \n",
       "max                 142.297333               116.161095   \n",
       "\n",
       "       cocoa_fung_hi_lab_usdha  cocoa_weed_hi_lab_usdha  \\\n",
       "count               963.000000               732.000000   \n",
       "mean                  3.686724                35.093170   \n",
       "std                  11.614023                54.913971   \n",
       "min                   0.000000                 0.000000   \n",
       "25%                   0.000000                 0.000000   \n",
       "50%                   0.000000                 4.980000   \n",
       "75%                   0.000000                50.590759   \n",
       "max                 103.254311               322.669739   \n",
       "\n",
       "       cocoa_pruning_hi_lab_usdha  cocoa_harvest_hi_lab_usdha  \\\n",
       "count                  598.000000                  662.000000   \n",
       "mean                     7.842022                   13.991180   \n",
       "std                     18.787256                   30.327866   \n",
       "min                      0.000000                    0.000000   \n",
       "25%                      0.000000                    0.000000   \n",
       "50%                      0.000000                    0.000000   \n",
       "75%                      0.000000                   15.488146   \n",
       "max                    125.406960                  206.508606   \n",
       "\n",
       "       cocoa_podbreak_hi_lab_usdha  cocoa_transport_hi_lab_usdha  \n",
       "count                   641.000000                    932.000000  \n",
       "mean                      1.790963                      4.574852  \n",
       "std                       8.157135                     12.273155  \n",
       "min                       0.000000                      0.000000  \n",
       "25%                       0.000000                      0.000000  \n",
       "50%                       0.000000                      0.000000  \n",
       "75%                       0.000000                      2.988000  \n",
       "max                      82.280777                    103.254303  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.filter(regex = 'hi_lab_usdha').describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As expected these features have missing values, in different levels. As the previous studies explain, this happens when farmers know they did some activity, but did not know the exact costs. \n",
    "\n",
    "Let's verify:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "yes    221\n",
       "no       5\n",
       "Name: cocoa_landclear_yn, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.loc[data['cocoa_landclear_hi_lab_usdha'].isnull(),'cocoa_landclear_yn'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "My strategy will be to:\n",
    "\n",
    "- enforce 0 if activity was not done\n",
    "- if the activity was done but it is missing, fill the missing values with the median values (per ha) of the farmers that did the activity and could specify the costs. I will do this per analytical group."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "def fill_na_by_group(df, yn_feature, feature, group_feature, strategy = 'median'):\n",
    "    '''\n",
    "    fills missing values in given feature with the group median (default) or mean value\n",
    "        \n",
    "    \n",
    "    INPUT:\n",
    "        df: panda DataFrame\n",
    "        yn_feature: colname to be used to slice the dataframe into doing (yes) or not doing (no) an activity\n",
    "        feature: colname to be filled\n",
    "        group_feature: feature identifying the different groups\n",
    "        strategy = 'median' or 'mean'\n",
    "        \n",
    "    OUTPUT:\n",
    "        df: modified DataFrame\n",
    "    '''\n",
    "    \n",
    "    # first enforce 0 if yn_feature = 'no'\n",
    "    df.loc[df[yn_feature]=='no', feature] = 0\n",
    "\n",
    "    # get a list of groups\n",
    "    groups = df[group_feature].unique()\n",
    "    \n",
    "    # for each group, filter on yn_feature = yes and fill in median\n",
    "    for group in groups:\n",
    "        \n",
    "        if strategy == 'median':\n",
    "            fill_value = df.loc[(df[group_feature] == group) & (df[yn_feature] == 'yes'), feature ].median()\n",
    "        elif strategy == 'mean':\n",
    "            fill_value = df.loc[(df[group_feature] == group) & (df[yn_feature] == 'yes'), feature ].mean()\n",
    "        else:\n",
    "            raise SyntaxError('Strategy unknown')\n",
    "        \n",
    "        df.loc[(df[group_feature] == group) & (df[yn_feature] == 'yes'), feature ] = df.loc[(df[group_feature] == group) & (df[yn_feature] == 'yes'), feature ].fillna(fill_value)\n",
    "       \n",
    "    \n",
    "    # return modified DataFrame\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# iterate through all types of labor\n",
    "for labor_type in types_of_labor:\n",
    "    \n",
    "    # find all columns with that labor type\n",
    "    full_list = list(data.filter(regex = labor_type))\n",
    "    \n",
    "    # apply the fill_na_by_group to each of these columns\n",
    "    for col in full_list:\n",
    "        yn_feature = col.replace(labor_type,'yn')\n",
    "        data = fill_na_by_group(data, yn_feature, col, 'Analytical Group')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check to be sure it worked as expected:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>cocoa_landclear_hi_lab_usdha</th>\n",
       "      <th>cocoa_landprep_hi_lab_usdha</th>\n",
       "      <th>cocoa_planting_hi_lab_usdha</th>\n",
       "      <th>cocoa_gfert_hi_lab_usdha</th>\n",
       "      <th>cocoa_lfert_hi_lab_usdha</th>\n",
       "      <th>cocoa_manurecomp_hi_lab_usdha</th>\n",
       "      <th>cocoa_herb_hi_lab_usdha</th>\n",
       "      <th>cocoa_pest_hi_lab_usdha</th>\n",
       "      <th>cocoa_fung_hi_lab_usdha</th>\n",
       "      <th>cocoa_weed_hi_lab_usdha</th>\n",
       "      <th>cocoa_pruning_hi_lab_usdha</th>\n",
       "      <th>cocoa_harvest_hi_lab_usdha</th>\n",
       "      <th>cocoa_podbreak_hi_lab_usdha</th>\n",
       "      <th>cocoa_transport_hi_lab_usdha</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1432.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>6.023585</td>\n",
       "      <td>1.798956</td>\n",
       "      <td>0.890675</td>\n",
       "      <td>0.748395</td>\n",
       "      <td>4.264726</td>\n",
       "      <td>0.225778</td>\n",
       "      <td>2.661115</td>\n",
       "      <td>9.449956</td>\n",
       "      <td>3.367711</td>\n",
       "      <td>32.008888</td>\n",
       "      <td>4.094665</td>\n",
       "      <td>8.575407</td>\n",
       "      <td>0.797227</td>\n",
       "      <td>3.174717</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>13.587420</td>\n",
       "      <td>9.865560</td>\n",
       "      <td>5.822273</td>\n",
       "      <td>4.069307</td>\n",
       "      <td>11.274717</td>\n",
       "      <td>2.142114</td>\n",
       "      <td>11.790038</td>\n",
       "      <td>14.716000</td>\n",
       "      <td>9.837022</td>\n",
       "      <td>41.412823</td>\n",
       "      <td>13.042006</td>\n",
       "      <td>21.667908</td>\n",
       "      <td>5.512402</td>\n",
       "      <td>10.076999</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>7.375307</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>29.685612</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>7.407631</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.338855</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>11.388343</td>\n",
       "      <td>1.289464</td>\n",
       "      <td>31.099216</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.717306</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.872825</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>129.067871</td>\n",
       "      <td>160.617813</td>\n",
       "      <td>90.347519</td>\n",
       "      <td>51.627155</td>\n",
       "      <td>99.382263</td>\n",
       "      <td>38.720364</td>\n",
       "      <td>142.297333</td>\n",
       "      <td>116.161095</td>\n",
       "      <td>103.254311</td>\n",
       "      <td>322.669739</td>\n",
       "      <td>125.406960</td>\n",
       "      <td>206.508606</td>\n",
       "      <td>82.280777</td>\n",
       "      <td>103.254303</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       cocoa_landclear_hi_lab_usdha  cocoa_landprep_hi_lab_usdha  \\\n",
       "count                   1440.000000                  1440.000000   \n",
       "mean                       6.023585                     1.798956   \n",
       "std                       13.587420                     9.865560   \n",
       "min                        0.000000                     0.000000   \n",
       "25%                        0.000000                     0.000000   \n",
       "50%                        0.000000                     0.000000   \n",
       "75%                        7.407631                     0.000000   \n",
       "max                      129.067871                   160.617813   \n",
       "\n",
       "       cocoa_planting_hi_lab_usdha  cocoa_gfert_hi_lab_usdha  \\\n",
       "count                  1440.000000               1440.000000   \n",
       "mean                      0.890675                  0.748395   \n",
       "std                       5.822273                  4.069307   \n",
       "min                       0.000000                  0.000000   \n",
       "25%                       0.000000                  0.000000   \n",
       "50%                       0.000000                  0.000000   \n",
       "75%                       0.000000                  0.000000   \n",
       "max                      90.347519                 51.627155   \n",
       "\n",
       "       cocoa_lfert_hi_lab_usdha  cocoa_manurecomp_hi_lab_usdha  \\\n",
       "count               1440.000000                    1432.000000   \n",
       "mean                   4.264726                       0.225778   \n",
       "std                   11.274717                       2.142114   \n",
       "min                    0.000000                       0.000000   \n",
       "25%                    0.000000                       0.000000   \n",
       "50%                    0.000000                       0.000000   \n",
       "75%                    3.338855                       0.000000   \n",
       "max                   99.382263                      38.720364   \n",
       "\n",
       "       cocoa_herb_hi_lab_usdha  cocoa_pest_hi_lab_usdha  \\\n",
       "count              1440.000000              1440.000000   \n",
       "mean                  2.661115                 9.449956   \n",
       "std                  11.790038                14.716000   \n",
       "min                   0.000000                 0.000000   \n",
       "25%                   0.000000                 0.000000   \n",
       "50%                   0.000000                 7.375307   \n",
       "75%                   0.000000                11.388343   \n",
       "max                 142.297333               116.161095   \n",
       "\n",
       "       cocoa_fung_hi_lab_usdha  cocoa_weed_hi_lab_usdha  \\\n",
       "count              1440.000000              1440.000000   \n",
       "mean                  3.367711                32.008888   \n",
       "std                   9.837022                41.412823   \n",
       "min                   0.000000                 0.000000   \n",
       "25%                   0.000000                 0.000000   \n",
       "50%                   0.000000                29.685612   \n",
       "75%                   1.289464                31.099216   \n",
       "max                 103.254311               322.669739   \n",
       "\n",
       "       cocoa_pruning_hi_lab_usdha  cocoa_harvest_hi_lab_usdha  \\\n",
       "count                 1440.000000                 1440.000000   \n",
       "mean                     4.094665                    8.575407   \n",
       "std                     13.042006                   21.667908   \n",
       "min                      0.000000                    0.000000   \n",
       "25%                      0.000000                    0.000000   \n",
       "50%                      0.000000                    0.000000   \n",
       "75%                      0.000000                    8.717306   \n",
       "max                    125.406960                  206.508606   \n",
       "\n",
       "       cocoa_podbreak_hi_lab_usdha  cocoa_transport_hi_lab_usdha  \n",
       "count                  1440.000000                   1440.000000  \n",
       "mean                      0.797227                      3.174717  \n",
       "std                       5.512402                     10.076999  \n",
       "min                       0.000000                      0.000000  \n",
       "25%                       0.000000                      0.000000  \n",
       "50%                       0.000000                      0.000000  \n",
       "75%                       0.000000                      1.872825  \n",
       "max                      82.280777                    103.254303  "
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.filter(regex = 'hi_lab_usdha').describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I can now calculate the total hired labor costs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['Cocoa hired labour costs (USD/ha)'] = data.filter(regex = 'hi_lab_usdha').sum(axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I use a similar strategy for the input costs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>cocoa_gfert_usdha</th>\n",
       "      <th>cocoa_lfert_usdha</th>\n",
       "      <th>cocoa_herb_usdha</th>\n",
       "      <th>cocoa_pest_usdha</th>\n",
       "      <th>cocoa_fung_usdha</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1416.000000</td>\n",
       "      <td>1405.000000</td>\n",
       "      <td>1363.000000</td>\n",
       "      <td>1348.000000</td>\n",
       "      <td>1369.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>8.633907</td>\n",
       "      <td>2.848809</td>\n",
       "      <td>7.123010</td>\n",
       "      <td>17.830339</td>\n",
       "      <td>2.592730</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>35.850471</td>\n",
       "      <td>9.368103</td>\n",
       "      <td>11.274213</td>\n",
       "      <td>22.563593</td>\n",
       "      <td>6.736555</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>10.755658</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12.019188</td>\n",
       "      <td>25.813578</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>464.644409</td>\n",
       "      <td>77.440727</td>\n",
       "      <td>67.115295</td>\n",
       "      <td>136.660110</td>\n",
       "      <td>53.119999</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       cocoa_gfert_usdha  cocoa_lfert_usdha  cocoa_herb_usdha  \\\n",
       "count        1416.000000        1405.000000       1363.000000   \n",
       "mean            8.633907           2.848809          7.123010   \n",
       "std            35.850471           9.368103         11.274213   \n",
       "min             0.000000           0.000000          0.000000   \n",
       "25%             0.000000           0.000000          0.000000   \n",
       "50%             0.000000           0.000000          0.000000   \n",
       "75%             0.000000           0.000000         12.019188   \n",
       "max           464.644409          77.440727         67.115295   \n",
       "\n",
       "       cocoa_pest_usdha  cocoa_fung_usdha  \n",
       "count       1348.000000       1369.000000  \n",
       "mean          17.830339          2.592730  \n",
       "std           22.563593          6.736555  \n",
       "min            0.000000          0.000000  \n",
       "25%            0.000000          0.000000  \n",
       "50%           10.755658          0.000000  \n",
       "75%           25.813578          0.000000  \n",
       "max          136.660110         53.119999  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "inputcosts_extended_list = ['cocoa_' + this_input + '_usdha' for this_input in inputs]\n",
    "data[inputcosts_extended_list].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# iterate through all input costs\n",
    "for col in inputcosts_extended_list:\n",
    "    yn_feature = col.replace(\"usdha\",'yn')\n",
    "    data = fill_na_by_group(data, yn_feature, col, 'Analytical Group')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Check"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>cocoa_gfert_usdha</th>\n",
       "      <th>cocoa_lfert_usdha</th>\n",
       "      <th>cocoa_herb_usdha</th>\n",
       "      <th>cocoa_pest_usdha</th>\n",
       "      <th>cocoa_fung_usdha</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "      <td>1440.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>8.719259</td>\n",
       "      <td>2.798434</td>\n",
       "      <td>7.551857</td>\n",
       "      <td>17.669798</td>\n",
       "      <td>2.471631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>35.630550</td>\n",
       "      <td>9.273187</td>\n",
       "      <td>11.145175</td>\n",
       "      <td>21.843811</td>\n",
       "      <td>6.591352</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12.177862</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>13.552127</td>\n",
       "      <td>24.879286</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>464.644409</td>\n",
       "      <td>77.440727</td>\n",
       "      <td>67.115295</td>\n",
       "      <td>136.660110</td>\n",
       "      <td>53.119999</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       cocoa_gfert_usdha  cocoa_lfert_usdha  cocoa_herb_usdha  \\\n",
       "count        1440.000000        1440.000000       1440.000000   \n",
       "mean            8.719259           2.798434          7.551857   \n",
       "std            35.630550           9.273187         11.145175   \n",
       "min             0.000000           0.000000          0.000000   \n",
       "25%             0.000000           0.000000          0.000000   \n",
       "50%             0.000000           0.000000          0.000000   \n",
       "75%             0.000000           0.000000         13.552127   \n",
       "max           464.644409          77.440727         67.115295   \n",
       "\n",
       "       cocoa_pest_usdha  cocoa_fung_usdha  \n",
       "count       1440.000000       1440.000000  \n",
       "mean          17.669798          2.471631  \n",
       "std           21.843811          6.591352  \n",
       "min            0.000000          0.000000  \n",
       "25%            0.000000          0.000000  \n",
       "50%           12.177862          0.000000  \n",
       "75%           24.879286          0.000000  \n",
       "max          136.660110         53.119999  "
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data[inputcosts_extended_list].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now I can also calculate the total input costs:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['Cocoa input costs (USD/ha)'] = data[inputcosts_extended_list].sum(axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.3 Value adjustments over time"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since a main goal of this analysis is to compare total incomes with the Living Income Benchmark value, they need to be comparable in purchasing power. The Living Income Benchmark values refer to Q1 2018 (Ghana) and Q2 2018 (Côte d'Ivoire). This data, on the other hand, refers to Q1 2016. Therefore I will use a [CPI correction factor](http://data.imf.org/regular.aspx?key=61545849) to bring the survey data from 2016 to 2018 values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# CPI Correction\n",
    "\n",
    "# Ghana CPI in 2018 Q1: 247.05\n",
    "# Ghana CPI in 2016 Q1: 197.77\n",
    "# Living Income  CEDI to USD = 1/4.45 // March 1st 2018\n",
    "# Survey data CEDI to USD = 0.26116 // Jan 15th, 2016\n",
    "correction_factor_ghana = (1/0.26116)/197.77*247.05/4.45\n",
    "\n",
    "\n",
    "# CdI CPI in 2018 Q2: 112.92\n",
    "# CdI CPI in 2016 Q1: 111.35\n",
    "# Living Income CFA to USD = 1/576.81 // August 2018\n",
    "# Survey data CFA to USD = 0.00166 // Jan 16th, 2016\n",
    "correction_factor_cdi = (1/0.00166)/111.35*112.92/576.81\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Apply the correction factor all variables in USD:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "usd_list = data.filter(regex = \"usd\").columns\n",
    "usd_list = usd_list.append(data.filter(regex = \"USD\").columns)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[data['country'] == 'Ghana', usd_list] = data.loc[data['country'] == 'Ghana', usd_list]*correction_factor_ghana\n",
    "data.loc[data['country'] == 'CdI', usd_list] = data.loc[data['country'] == 'CdI', usd_list]*correction_factor_cdi"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.4 Total household income estimation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Finally I will check if all households have estimated share of income coming from cocoa:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    1438.000000\n",
       "mean        0.628803\n",
       "std         0.214800\n",
       "min         0.000000\n",
       "25%         0.500000\n",
       "50%         0.600000\n",
       "75%         0.800000\n",
       "max         1.000000\n",
       "Name: hh_income_perc_cocoa, dtype: float64"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['hh_income_perc_cocoa'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['hh_income_perc_cocoa'].isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I will also fill these missing values with the group median"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['temp']='yes'\n",
    "data = fill_na_by_group(data,'temp', 'hh_income_perc_cocoa', 'Analytical Group')\n",
    "data.drop('temp', axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I also need to drop observations where the share is 0%"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data[data['hh_income_perc_cocoa']>0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now I can compute total household income:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['Estimated total household income (USD/household/year)'] = (\n",
    "    (\n",
    "        + data['cocoa_prod_total_kgsha']*data['cocoa_price_usdkg'] \n",
    "        - data['Cocoa input costs (USD/ha)']\n",
    "        - data['Cocoa hired labour costs (USD/ha)']\n",
    "    )*data['cocoa_land_used_morethan5_ha'] \n",
    "    /data['hh_income_perc_cocoa']\n",
    ")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's check:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count     1434.000000\n",
       "mean      3165.165283\n",
       "std       3846.370605\n",
       "min     -16244.933594\n",
       "25%        989.524460\n",
       "50%       2002.150330\n",
       "75%       3960.125427\n",
       "max      49948.113281\n",
       "Name: Estimated total household income (USD/household/year), dtype: float64"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Estimated total household income (USD/household/year)'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's drop observations where the total household income is smaller than 0 (likely due to erros in the other variables) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = data[data['Estimated total household income (USD/household/year)']>0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's also drop observations that have income values more than 4 standard deviations above the mean, in a iterative loop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "upper_limit = data['Estimated total household income (USD/household/year)'].mean() + data['Estimated total household income (USD/household/year)'].std()*4\n",
    "\n",
    "while data['Estimated total household income (USD/household/year)'].max() > upper_limit:\n",
    "    data = data[data['Estimated total household income (USD/household/year)']<=upper_limit]\n",
    "    upper_limit = data['Estimated total household income (USD/household/year)'].mean() + data['Estimated total household income (USD/household/year)'].std()*4"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1.5 Exploration of total household income"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's check basic descriptives:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Ghana</th>\n",
       "      <td>970.0</td>\n",
       "      <td>2821.449707</td>\n",
       "      <td>2594.181152</td>\n",
       "      <td>15.572671</td>\n",
       "      <td>1052.486908</td>\n",
       "      <td>1910.728027</td>\n",
       "      <td>3773.894104</td>\n",
       "      <td>13381.899414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CdI</th>\n",
       "      <td>407.0</td>\n",
       "      <td>2978.242432</td>\n",
       "      <td>2723.840088</td>\n",
       "      <td>18.753244</td>\n",
       "      <td>996.925751</td>\n",
       "      <td>2158.178467</td>\n",
       "      <td>3966.554443</td>\n",
       "      <td>13298.229492</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         count         mean          std        min          25%          50%  \\\n",
       "country                                                                         \n",
       "Ghana    970.0  2821.449707  2594.181152  15.572671  1052.486908  1910.728027   \n",
       "CdI      407.0  2978.242432  2723.840088  18.753244   996.925751  2158.178467   \n",
       "\n",
       "                 75%           max  \n",
       "country                             \n",
       "Ghana    3773.894104  13381.899414  \n",
       "CdI      3966.554443  13298.229492  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.groupby('country')['Estimated total household income (USD/household/year)'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Analytical Group</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>CdI: Male-headed, large</th>\n",
       "      <td>101.0</td>\n",
       "      <td>5046.209961</td>\n",
       "      <td>3238.920166</td>\n",
       "      <td>387.704834</td>\n",
       "      <td>2573.553955</td>\n",
       "      <td>4286.904785</td>\n",
       "      <td>7436.996094</td>\n",
       "      <td>13298.229492</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CdI: Male-headed, typical</th>\n",
       "      <td>306.0</td>\n",
       "      <td>2295.678223</td>\n",
       "      <td>2133.793701</td>\n",
       "      <td>18.753244</td>\n",
       "      <td>855.932037</td>\n",
       "      <td>1751.853882</td>\n",
       "      <td>2847.878174</td>\n",
       "      <td>11328.135742</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ghana: Female-headed</th>\n",
       "      <td>150.0</td>\n",
       "      <td>1679.433350</td>\n",
       "      <td>1760.737671</td>\n",
       "      <td>15.572671</td>\n",
       "      <td>534.136765</td>\n",
       "      <td>1039.841949</td>\n",
       "      <td>2080.295502</td>\n",
       "      <td>9636.550781</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ghana: Male-headed, large</th>\n",
       "      <td>204.0</td>\n",
       "      <td>5280.251465</td>\n",
       "      <td>3382.007324</td>\n",
       "      <td>281.954163</td>\n",
       "      <td>2531.117065</td>\n",
       "      <td>4748.681641</td>\n",
       "      <td>7601.145386</td>\n",
       "      <td>13381.899414</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ghana: Male-headed, typical</th>\n",
       "      <td>616.0</td>\n",
       "      <td>2285.259033</td>\n",
       "      <td>1861.575562</td>\n",
       "      <td>68.563194</td>\n",
       "      <td>1052.614532</td>\n",
       "      <td>1721.023865</td>\n",
       "      <td>2985.473633</td>\n",
       "      <td>12828.942383</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             count         mean          std         min  \\\n",
       "Analytical Group                                                           \n",
       "CdI: Male-headed, large      101.0  5046.209961  3238.920166  387.704834   \n",
       "CdI: Male-headed, typical    306.0  2295.678223  2133.793701   18.753244   \n",
       "Ghana: Female-headed         150.0  1679.433350  1760.737671   15.572671   \n",
       "Ghana: Male-headed, large    204.0  5280.251465  3382.007324  281.954163   \n",
       "Ghana: Male-headed, typical  616.0  2285.259033  1861.575562   68.563194   \n",
       "\n",
       "                                     25%          50%          75%  \\\n",
       "Analytical Group                                                     \n",
       "CdI: Male-headed, large      2573.553955  4286.904785  7436.996094   \n",
       "CdI: Male-headed, typical     855.932037  1751.853882  2847.878174   \n",
       "Ghana: Female-headed          534.136765  1039.841949  2080.295502   \n",
       "Ghana: Male-headed, large    2531.117065  4748.681641  7601.145386   \n",
       "Ghana: Male-headed, typical  1052.614532  1721.023865  2985.473633   \n",
       "\n",
       "                                      max  \n",
       "Analytical Group                           \n",
       "CdI: Male-headed, large      13298.229492  \n",
       "CdI: Male-headed, typical    11328.135742  \n",
       "Ghana: Female-headed          9636.550781  \n",
       "Ghana: Male-headed, large    13381.899414  \n",
       "Ghana: Male-headed, typical  12828.942383  "
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.groupby('Analytical Group')['Estimated total household income (USD/household/year)'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lets plot some histograms:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "plt.title ('Ditribution of incomes')\n",
    "sns.distplot(data['Estimated total household income (USD/household/year)']);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "plt.title ('Ditribution of incomes per country')\n",
    "sns.kdeplot(data[data['country']=='Ghana']['Estimated total household income (USD/household/year)'],  shade=True , label = 'Ghana');\n",
    "sns.kdeplot(data[data['country']=='CdI']['Estimated total household income (USD/household/year)'], shade = True, label = 'Côte d\\'Ivoire');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "plt.title ('Ditribution of incomes in Ghana')\n",
    "plt.xlabel(\"Household Annual Income (USD)\")\n",
    "sns.kdeplot(data[data['Analytical Group']=='Ghana: Female-headed']['Estimated total household income (USD/household/year)'], shade = True, label = 'Female-headed');\n",
    "sns.kdeplot(data[data['Analytical Group']=='Ghana: Male-headed, typical']['Estimated total household income (USD/household/year)'], shade = True, label = 'Male-headed, typical')\n",
    "sns.kdeplot(data[data['Analytical Group']=='Ghana: Male-headed, large']['Estimated total household income (USD/household/year)'], shade = True, label = 'Male-headed, large');\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "plt.title ('Ditribution of incomes in Cotê d\\'Ivoire')\n",
    "plt.xlabel(\"Household Annual Income (USD)\")\n",
    "sns.kdeplot(data[data['Analytical Group']=='CdI: Male-headed, typical']['Estimated total household income (USD/household/year)'], shade = True, label = 'Male-headed, typical')\n",
    "sns.kdeplot(data[data['Analytical Group']=='CdI: Male-headed, large']['Estimated total household income (USD/household/year)'], shade = True, label = 'Male-headed, large');\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2 Living Income Benchmark"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I will create a feature including the living benchmark value and an indicator variable for achieving it. These values can be found  [here](https://c69aa8ac-6965-42b2-abb7-0f0b86c23d2e.filesusr.com/ugd/0c5ab3_fc3386a550b94a898c7757ee13ab59e6.pdf) and [here](https://c69aa8ac-6965-42b2-abb7-0f0b86c23d2e.filesusr.com/ugd/0c5ab3_93560a9b816d40c3a28daaa686e972a5.pdf)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Analytical Group</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>CdI: Male-headed, large</th>\n",
       "      <td>101.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>6518.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CdI: Male-headed, typical</th>\n",
       "      <td>306.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>6518.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ghana: Female-headed</th>\n",
       "      <td>150.0</td>\n",
       "      <td>4001.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4001.0</td>\n",
       "      <td>4001.0</td>\n",
       "      <td>4001.0</td>\n",
       "      <td>4001.0</td>\n",
       "      <td>4001.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ghana: Male-headed, large</th>\n",
       "      <td>204.0</td>\n",
       "      <td>5123.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5123.0</td>\n",
       "      <td>5123.0</td>\n",
       "      <td>5123.0</td>\n",
       "      <td>5123.0</td>\n",
       "      <td>5123.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ghana: Male-headed, typical</th>\n",
       "      <td>616.0</td>\n",
       "      <td>4742.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4742.0</td>\n",
       "      <td>4742.0</td>\n",
       "      <td>4742.0</td>\n",
       "      <td>4742.0</td>\n",
       "      <td>4742.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             count    mean  std     min     25%     50%  \\\n",
       "Analytical Group                                                          \n",
       "CdI: Male-headed, large      101.0  6518.0  0.0  6518.0  6518.0  6518.0   \n",
       "CdI: Male-headed, typical    306.0  6518.0  0.0  6518.0  6518.0  6518.0   \n",
       "Ghana: Female-headed         150.0  4001.0  0.0  4001.0  4001.0  4001.0   \n",
       "Ghana: Male-headed, large    204.0  5123.0  0.0  5123.0  5123.0  5123.0   \n",
       "Ghana: Male-headed, typical  616.0  4742.0  0.0  4742.0  4742.0  4742.0   \n",
       "\n",
       "                                75%     max  \n",
       "Analytical Group                             \n",
       "CdI: Male-headed, large      6518.0  6518.0  \n",
       "CdI: Male-headed, typical    6518.0  6518.0  \n",
       "Ghana: Female-headed         4001.0  4001.0  \n",
       "Ghana: Male-headed, large    5123.0  5123.0  \n",
       "Ghana: Male-headed, typical  4742.0  4742.0  "
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.loc[data['country']=='CdI','Living Income Benchmark'] = 6518\n",
    "data.loc[data['Analytical Group']=='Ghana: Female-headed','Living Income Benchmark'] = 4001\n",
    "data.loc[data['Analytical Group']=='Ghana: Male-headed, typical','Living Income Benchmark'] = 4742\n",
    "data.loc[data['Analytical Group']=='Ghana: Male-headed, large','Living Income Benchmark'] = 5123\n",
    "\n",
    "data.groupby('Analytical Group')['Living Income Benchmark'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Create the indicator"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['Living Income Achieved'] = data['Estimated total household income (USD/household/year)']>= data['Living Income Benchmark']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## QUESTION 1: How many farming households achieve a living Income?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "15% reach a living income\n"
     ]
    }
   ],
   "source": [
    "print(\"{0}% reach a living income\".format(round(data['Living Income Achieved'].mean()*100)))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Is quite confronting to realize that only this amount of farmers producing cocoa reach a living income"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## QUESTION 2: Is there a difference per group?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary = data.groupby('Analytical Group')['Living Income Achieved'].mean()*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Analytical Group\n",
       "CdI: Male-headed, large        27.722772\n",
       "CdI: Male-headed, typical       6.535948\n",
       "Ghana: Female-headed            9.333333\n",
       "Ghana: Male-headed, large      42.647059\n",
       "Ghana: Male-headed, typical     9.090909\n",
       "Name: Living Income Achieved, dtype: float64"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "summary"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's make a graph per group:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10))\n",
    "plt.ylabel(\"Percentage farming households reaching a living income\")\n",
    "plt.title(\"How many farmers reach a living income by group? (red line shows overall mean)\")\n",
    "ax = summary.plot.bar(yticks = range(0,110,10));\n",
    "ax.axhline(y=round(data['Living Income Achieved'].mean()*100),\n",
    "              color='r', linestyle='--', lw=2, label = \"abc\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is clearly a difference per group. Households with larger farms are much more likely to reach a living income"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## QUESTION 3: What is the gap of median household (per group)?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's create a summary dataframe with the median values:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary_median = data.groupby('Analytical Group')[\n",
    "    ['Estimated total household income (USD/household/year)', 'Living Income Benchmark']\n",
    "].median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "summary_median['Income Share'] = summary_median['Estimated total household income (USD/household/year)']/summary_median['Living Income Benchmark']*100\n",
    "summary_median['Gap Share'] = 100 - summary_median['Income Share']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Estimated total household income (USD/household/year)</th>\n",
       "      <th>Living Income Benchmark</th>\n",
       "      <th>Income Share</th>\n",
       "      <th>Gap Share</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Analytical Group</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>CdI: Male-headed, large</th>\n",
       "      <td>4286.904785</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>65.770248</td>\n",
       "      <td>34.229752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CdI: Male-headed, typical</th>\n",
       "      <td>1751.853882</td>\n",
       "      <td>6518.0</td>\n",
       "      <td>26.877169</td>\n",
       "      <td>73.122831</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ghana: Female-headed</th>\n",
       "      <td>1039.841919</td>\n",
       "      <td>4001.0</td>\n",
       "      <td>25.989551</td>\n",
       "      <td>74.010449</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ghana: Male-headed, large</th>\n",
       "      <td>4748.681641</td>\n",
       "      <td>5123.0</td>\n",
       "      <td>92.693376</td>\n",
       "      <td>7.306624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Ghana: Male-headed, typical</th>\n",
       "      <td>1721.023926</td>\n",
       "      <td>4742.0</td>\n",
       "      <td>36.293208</td>\n",
       "      <td>63.706792</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                             Estimated total household income (USD/household/year)  \\\n",
       "Analytical Group                                                                     \n",
       "CdI: Male-headed, large                                            4286.904785       \n",
       "CdI: Male-headed, typical                                          1751.853882       \n",
       "Ghana: Female-headed                                               1039.841919       \n",
       "Ghana: Male-headed, large                                          4748.681641       \n",
       "Ghana: Male-headed, typical                                        1721.023926       \n",
       "\n",
       "                             Living Income Benchmark  Income Share  Gap Share  \n",
       "Analytical Group                                                               \n",
       "CdI: Male-headed, large                       6518.0     65.770248  34.229752  \n",
       "CdI: Male-headed, typical                     6518.0     26.877169  73.122831  \n",
       "Ghana: Female-headed                          4001.0     25.989551  74.010449  \n",
       "Ghana: Male-headed, large                     5123.0     92.693376   7.306624  \n",
       "Ghana: Male-headed, typical                   4742.0     36.293208  63.706792  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "summary_median"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's visualize this as stacked bar chart:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "summary_median[['Income Share', 'Gap Share']].plot.bar(stacked=True, figsize=(10,10),\n",
    "                                                      ylabel =\"Percentage of the Living Income benchmark\",\n",
    "                                                      yticks = range(0,110,10)\n",
    "                                                      );"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I can see that there is a similar difference per group, but even the group that mostly reach the living income, there is still a gap of 7% in the median household."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 3 Prepare for Machine Learning"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now I want to buld a classifier that will predict if a farming household reaches or not the Living Income Benchmark.\n",
    "\n",
    "The goal here is to have a model with as few variable as possible, which are likely to be either know by companies working with these farmers or relatevely easy to collect."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The creation of the classifier is show in [this](LivingIncome_MachineLearning.ipynb) jupyter notebook (for easiness of reading), but I'll keep the data preparation here."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.1 Feature Selection"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I'll further select features by removing features that should not be used by the ML model or that have too many missing values to be used."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The goal is to predict if a farming household achieves the Living Income Benchmark, without knowing their total household income (if one would know the total household income, or be able to compute, there is no need for a classifier model).\n",
    "\n",
    "There I'll drop all the calculated income variables I just built, plus the agregated and detailed costs of labour and inputs. I'll keep only indicator features of activities and sex and type of labor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop([\n",
    "    'Cocoa hired labour costs (USD/ha)',\n",
    "    'Cocoa input costs (USD/ha)',\n",
    "    'Estimated total household income (USD/household/year)',\n",
    "    'Living Income Benchmark',\n",
    "    'cocoa_marketing_income_usd',\n",
    "    'cocoa_marketing_income_usdha'\n",
    "], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop([\n",
    " 'cocoa_landclear_hi_n_ha',\n",
    " 'cocoa_landclear_hh_n_ha',\n",
    " 'cocoa_landclear_com_n_ha',\n",
    " 'cocoa_landclear_lab_n_ha',\n",
    " 'cocoa_landclear_hi_lab_usdha',\n",
    " 'cocoa_landprep_hi_n_ha',\n",
    " 'cocoa_landprep_hh_n_ha',\n",
    " 'cocoa_landprep_com_n_ha',\n",
    " 'cocoa_landprep_lab_n_ha',\n",
    " 'cocoa_landprep_hi_lab_usdha',\n",
    " 'cocoa_planting_hi_n_ha',\n",
    " 'cocoa_planting_hh_n_ha',\n",
    " 'cocoa_planting_com_n_ha',\n",
    " 'cocoa_planting_lab_n_ha',\n",
    " 'cocoa_planting_hi_lab_usdha',\n",
    " 'cocoa_gfert_hi_n_ha',\n",
    " 'cocoa_gfert_hh_n_ha',\n",
    " 'cocoa_gfert_com_n_ha',\n",
    " 'cocoa_gfert_lab_n_ha',\n",
    " 'cocoa_gfert_hi_lab_usdha',\n",
    " 'cocoa_lfert_hi_n_ha',\n",
    " 'cocoa_lfert_hh_n_ha',\n",
    " 'cocoa_lfert_com_n_ha',\n",
    " 'cocoa_lfert_lab_n_ha',\n",
    " 'cocoa_lfert_hi_lab_usdha',\n",
    " 'cocoa_manurecomp_hi_n_ha',\n",
    " 'cocoa_manurecomp_hh_n_ha',\n",
    " 'cocoa_manurecomp_com_n_ha',\n",
    " 'cocoa_manurecomp_lab_n_ha',\n",
    " 'cocoa_manurecomp_hi_lab_usdha',\n",
    " 'cocoa_herb_hi_n_ha',\n",
    " 'cocoa_herb_hh_n_ha',\n",
    " 'cocoa_herb_com_n_ha',\n",
    " 'cocoa_herb_lab_n_ha',\n",
    " 'cocoa_herb_hi_lab_usdha',\n",
    " 'cocoa_pest_hi_n_ha',\n",
    " 'cocoa_pest_hh_n_ha',\n",
    " 'cocoa_pest_com_n_ha',\n",
    " 'cocoa_pest_lab_n_ha',\n",
    " 'cocoa_pest_hi_lab_usdha',\n",
    " 'cocoa_fung_hi_n_ha',\n",
    " 'cocoa_fung_hh_n_ha',\n",
    " 'cocoa_fung_com_n_ha',\n",
    " 'cocoa_fung_lab_n_ha',\n",
    " 'cocoa_fung_hi_lab_usdha',\n",
    " 'cocoa_weed_hi_n_ha',\n",
    " 'cocoa_weed_hh_n_ha',\n",
    " 'cocoa_weed_com_n_ha',\n",
    " 'cocoa_weed_lab_n_ha',\n",
    " 'cocoa_weed_hi_lab_usdha',\n",
    " 'cocoa_pruning_hi_n_ha',\n",
    " 'cocoa_pruning_hh_n_ha',\n",
    " 'cocoa_pruning_com_n_ha',\n",
    " 'cocoa_pruning_lab_n_ha',\n",
    " 'cocoa_pruning_hi_lab_usdha',\n",
    " 'cocoa_harvest_hi_n_ha',\n",
    " 'cocoa_harvest_hh_n_ha',\n",
    " 'cocoa_harvest_com_n_ha',\n",
    " 'cocoa_harvest_lab_n_ha',\n",
    " 'cocoa_harvest_hi_lab_usdha',\n",
    " 'cocoa_podbreak_hi_n_ha',\n",
    " 'cocoa_podbreak_hh_n_ha',\n",
    " 'cocoa_podbreak_com_n_ha',\n",
    " 'cocoa_podbreak_lab_n_ha',\n",
    " 'cocoa_podbreak_hi_lab_usdha',\n",
    " 'cocoa_transport_hi_n_ha',\n",
    " 'cocoa_transport_hh_n_ha',\n",
    " 'cocoa_transport_com_n_ha',\n",
    " 'cocoa_transport_lab_n_ha',\n",
    " 'cocoa_transport_hi_lab_usdha',\n",
    " 'cocoa_landclear_times',\n",
    " 'cocoa_landclear_ha',\n",
    " 'cocoa_landclear_hi_cost',\n",
    " 'cocoa_landprep_times',\n",
    " 'cocoa_landprep_ha',\n",
    " 'cocoa_landprep_how_1',\n",
    " 'cocoa_landprep_how_2',\n",
    " 'cocoa_landprep_how_3',\n",
    " 'cocoa_landprep_how_4',\n",
    " 'cocoa_landprep_hi_cost',\n",
    " 'cocoa_planting_times',\n",
    " 'cocoa_planting_ha',\n",
    " 'cocoa_planting_hi_cost',\n",
    " 'cocoa_gfert_hi_cost',\n",
    " 'cocoa_gfert_times',\n",
    " 'cocoa_gfert_given_yn',\n",
    " 'cocoa_gfert_purchased_yn',\n",
    " 'cocoa_gfert_given_pack',\n",
    " 'cocoa_gfert_purchased_pack',\n",
    " 'cocoa_gfert_pack_used_total',\n",
    " 'cocoa_gfert_given_unitha',\n",
    " 'cocoa_gfert_purchased_unitha',\n",
    " 'cocoa_gfert_applied_unitha',\n",
    " 'cocoa_gfert_usdha',\n",
    " 'cocoa_lfert_times',\n",
    " 'cocoa_lfert_given_yn',\n",
    " 'cocoa_lfert_purchased_yn',\n",
    " 'cocoa_lfert_given_pack',\n",
    " 'cocoa_lfert_purchased_pack',\n",
    " 'cocoa_lfert_pack_used_total',\n",
    " 'cocoa_lfert_given_unitha',\n",
    " 'cocoa_lfert_purchased_unitha',\n",
    " 'cocoa_lfert_applied_unitha',\n",
    " 'cocoa_lfert_usdha',\n",
    " 'cocoa_lfert_hi_cost',\n",
    " 'cocoa_manurecomp_times',\n",
    " 'cocoa_manurecomp_hi_cost',\n",
    " 'cocoa_herb_times',\n",
    " 'cocoa_herb_given_yn',\n",
    " 'cocoa_herb_purchased_yn',\n",
    " 'cocoa_herb_given_pack',\n",
    " 'cocoa_herb_purchased_pack',\n",
    " 'cocoa_herb_pack_used_total',\n",
    " 'cocoa_herb_given_unitha',\n",
    " 'cocoa_herb_purchased_unitha',\n",
    " 'cocoa_herb_applied_unitha',\n",
    " 'cocoa_herb_usdha',\n",
    " 'cocoa_herb_hi_cost',\n",
    " 'cocoa_pest_times',\n",
    " 'cocoa_pest_given_yn',\n",
    " 'cocoa_pest_purchased_yn',\n",
    " 'cocoa_pest_given_pack',\n",
    " 'cocoa_pest_purchased_pack',\n",
    " 'cocoa_pest_pack_used_total',\n",
    " 'cocoa_pest_given_unitha',\n",
    " 'cocoa_pest_purchased_unitha',\n",
    " 'cocoa_pest_applied_unitha',\n",
    " 'cocoa_pest_usdha',\n",
    " 'cocoa_pest_hi_cost',\n",
    " 'cocoa_fung_times',\n",
    " 'cocoa_fung_given_yn',\n",
    " 'cocoa_fung_purchased_yn',\n",
    " 'cocoa_fung_given_pack',\n",
    " 'cocoa_fung_purchased_pack',\n",
    " 'cocoa_fung_pack_used_total',\n",
    " 'cocoa_fung_given_unitha',\n",
    " 'cocoa_fung_purchased_unitha',\n",
    " 'cocoa_fung_applied_unitha',\n",
    " 'cocoa_fung_usdha',\n",
    " 'cocoa_fung_hi_cost',\n",
    " 'cocoa_weed_times',\n",
    " 'cocoa_weed_hi_cost',\n",
    " 'cocoa_pruning_times',\n",
    " 'cocoa_pruning_hi_cost',\n",
    " 'cocoa_harvest_times',\n",
    " 'cocoa_harvest_hi_cost',\n",
    " 'cocoa_podbreak_times',\n",
    " 'cocoa_podbreak_wait_days',\n",
    " 'cocoa_podbreak_hi_cost',\n",
    " 'cocoa_ferment_wait_days',\n",
    " 'cocoa_ferment_material_1',\n",
    " 'cocoa_ferment_material_2',\n",
    " 'cocoa_ferment_material_3',\n",
    " 'cocoa_ferment_material_4',\n",
    " 'cocoa_ferment_material_5',\n",
    " 'cocoa_ferment_material_999',\n",
    " 'cocoa_ferment_material_99',\n",
    " 'cocoa_transport_means',\n",
    " 'cocoa_transport_times',\n",
    " 'cocoa_transport_hi_cost',\n",
    " 'cocoa_dry_wait_days',\n",
    " 'cocoa_dry_material',\n",
    "], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lets also drop all `yes/no`, questions which are only filters in the survey"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(list(data.filter(regex = 'known_yn')), axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since these are all cocoa farmers, let's drop the question on most important crop"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's first check if there is indeed little variation:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Cocoa                   92.447349\n",
       "Cashews                  1.815541\n",
       "Rubber                   1.597676\n",
       "Palm                     1.016703\n",
       "Rice                     0.726216\n",
       "Plantain                 0.435730\n",
       "Cassava                  0.290487\n",
       "Maize                    0.290487\n",
       "none/ not applicable     0.217865\n",
       "Eggplant                 0.217865\n",
       "Tomatoes                 0.217865\n",
       "Pineapple                0.145243\n",
       "Ginger                   0.145243\n",
       "Chili                    0.072622\n",
       "Coconut                  0.072622\n",
       "Coffee                   0.072622\n",
       "Groundnuts               0.072622\n",
       "Other                    0.072622\n",
       "Yam                      0.072622\n",
       "Garlic                   0.000000\n",
       "Cocoyam                  0.000000\n",
       "Okra                     0.000000\n",
       "Oranges                  0.000000\n",
       "Papaya                   0.000000\n",
       "Peppers                  0.000000\n",
       "Cabbages                 0.000000\n",
       "Beans                    0.000000\n",
       "Avocado                  0.000000\n",
       "Name: crops_important1, dtype: float64"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['crops_important1'].value_counts()/len(data.index)*100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(['crops_important1', 'crops_important1_5ago', 'crops_important2_5ago'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3.2 Investigate Missing values more in detail"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I'll create a function to compute the number of missing values per variable, as a share of the total"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "def missing_per_variable(data, describe = True):\n",
    "    '''\n",
    "    computes the number of missing values per variable\n",
    "        \n",
    "    INPUT:\n",
    "        data: panda DataFrame\n",
    "        describe: print a decription table\n",
    "        \n",
    "    OUTPUT:\n",
    "        share_missing: panda Series with the share of missing values per variables\n",
    "    '''\n",
    "    share_missing = data.isnull().sum()/len(data.index)*100\n",
    "    \n",
    "    if describe:\n",
    "        print(share_missing.describe(percentiles = [.25, .5, .75, .90, .95, .99],))\n",
    "    \n",
    "    return share_missing"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I'll also create a function that plots the missing values per variable:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_missing(share_missing, cutoff = 5):\n",
    "    '''\n",
    "    plots the share of missing value per variable\n",
    "        \n",
    "    INPUT:\n",
    "        share_missing: panda series created with missing_per_variable()\n",
    "        cutoff: minimum share (0-100) to be included in the plot\n",
    "        \n",
    "    OUTPUT:\n",
    "        nothing (plots the graph on the notebook)\n",
    "    '''\n",
    "    \n",
    "    share_missing[share_missing>cutoff].sort_values().plot.bar(\n",
    "        figsize=(15,10), \n",
    "        color=(0.2, 0.4, 0.6, 0.6)\n",
    "    )\n",
    "    plt.title(\"Ordered barplot of missing values per feature\")\n",
    "    plt.ylabel(\"Share of missing values (%)\")\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count    689.000000\n",
      "mean      25.291514\n",
      "std       32.596694\n",
      "min        0.000000\n",
      "25%        0.000000\n",
      "50%        1.742919\n",
      "75%       55.700799\n",
      "90%       78.576616\n",
      "95%       91.430646\n",
      "99%       99.709513\n",
      "max       99.709513\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    " # check missing per variables:\n",
    "share_missing = missing_per_variable(data)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that there are still variables with many missing values. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's plot to confirm:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_missing(share_missing)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Even though the graph is not readable, it confirms that some variables have many missing values.\n",
    "\n",
    "Let's first take a look a the most problematic variables:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "117  problems\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "['all_land_leasedout_landunit_ha',\n",
       " 'cocoa_trees_num',\n",
       " 'cocoa_land_dec_how_5ago',\n",
       " 'cocoa_intercrop_25yo',\n",
       " 'cocoa_tree_spacing_trees',\n",
       " 'cocoa_tree_spacing_rows',\n",
       " 'cocoa_manurecomp_hhhiredcom_1',\n",
       " 'cocoa_manurecomp_hhhiredcom_2',\n",
       " 'cocoa_manurecomp_hhhiredcom_3',\n",
       " 'cocoa_manurecomp_hhhiredcom_4',\n",
       " 'cocoa_manurecomp_gender_1',\n",
       " 'cocoa_manurecomp_gender_2',\n",
       " 'cocoa_irrigation_source_1',\n",
       " 'cocoa_irrigation_source_2',\n",
       " 'cocoa_irrigation_source_3',\n",
       " 'cocoa_irrigation_source_4',\n",
       " 'cocoa_irrigation_source_5',\n",
       " 'cocoa_irrigation_source_6',\n",
       " 'cocoa_irrigation_source_7',\n",
       " 'cocoa_irrigation_method_1',\n",
       " 'cocoa_irrigation_method_2',\n",
       " 'cocoa_irrigation_method_3',\n",
       " 'cocoa_irrigation_method_4',\n",
       " 'cocoa_irrigation_method_5',\n",
       " 'cocoa_irrigation_method_6',\n",
       " 'cocoa_irrigation_method_7',\n",
       " 'cocoa_losses_storage_affect',\n",
       " 'cocoa_cert_type_clean_1',\n",
       " 'cocoa_cert_type_clean_2',\n",
       " 'cocoa_cert_type_clean_3',\n",
       " 'cocoa_cert_type_clean_4',\n",
       " 'cocoa_prodgroup_services_1',\n",
       " 'cocoa_prodgroup_services_2',\n",
       " 'cocoa_prodgroup_services_3',\n",
       " 'cocoa_prodgroup_services_4',\n",
       " 'cocoa_prodgroup_services_5',\n",
       " 'cocoa_prodgroup_services_6',\n",
       " 'cocoa_prodgroup_services_7',\n",
       " 'cocoa_prodgroup_services_8',\n",
       " 'cocoa_prodgroup_services_9',\n",
       " 'cocoa_prodgroup_services_10',\n",
       " 'cocoa_prodgroup_services_999',\n",
       " 'cocoa_credit_inputs_1',\n",
       " 'cocoa_credit_inputs_2',\n",
       " 'cocoa_credit_inputs_3',\n",
       " 'cocoa_credit_inputs_4',\n",
       " 'cocoa_credit_whom',\n",
       " 'loan_source_1',\n",
       " 'loan_source_2',\n",
       " 'loan_source_3',\n",
       " 'loan_source_4',\n",
       " 'loan_source_5',\n",
       " 'loan_source_6',\n",
       " 'loan_source_7',\n",
       " 'loan_source_8',\n",
       " 'loan_source_9',\n",
       " 'loan_source_10',\n",
       " 'loan_source_11',\n",
       " 'loan_source_12',\n",
       " 'loan_source_13',\n",
       " 'loan_source_999',\n",
       " 'loan_amount_usd',\n",
       " 'loan_repayment_success',\n",
       " 'loan_repayment_c',\n",
       " 'loan_use_1',\n",
       " 'loan_use_2',\n",
       " 'loan_use_3',\n",
       " 'loan_use_4',\n",
       " 'loan_use_5',\n",
       " 'loan_use_6',\n",
       " 'loan_use_7',\n",
       " 'loan_use_8',\n",
       " 'loan_use_9',\n",
       " 'loan_use_10',\n",
       " 'loan_use_11',\n",
       " 'loan_use_12',\n",
       " 'loan_use_13',\n",
       " 'loan_use_14',\n",
       " 'loan_use_15',\n",
       " 'loan_use_16',\n",
       " 'loan_use_17',\n",
       " 'loan_use_19',\n",
       " 'loan_use_20',\n",
       " 'loan_use_18',\n",
       " 'loan_use_999',\n",
       " 'hh_tractor',\n",
       " 'hh_cable_subscription',\n",
       " 'hh_fan',\n",
       " 'hh_gaselectric_stove',\n",
       " 'nutrition_a_grains',\n",
       " 'nutrition_b_roots_tubers_plant',\n",
       " 'nutrition_c_pulses',\n",
       " 'nutrition_d_nuts_seeds',\n",
       " 'nutrition_e_milk_products',\n",
       " 'nutrition_f_organ_meat',\n",
       " 'nutrition_g_meat_poultry',\n",
       " 'nutrition_h_fish_seafood',\n",
       " 'nutrition_i_eggs',\n",
       " 'nutrition_j_dark_leafy_veg',\n",
       " 'nutrition_k_vita_vegroots',\n",
       " 'nutrition_l_vita_fruits',\n",
       " 'nutrition_m_other_vegetables',\n",
       " 'nutrition_n_other_fruits',\n",
       " 'nutrition_condiments',\n",
       " 'nutrition_other_bevfood',\n",
       " 'MDD_1_W',\n",
       " 'MDD_2_W',\n",
       " 'MDD_3_W',\n",
       " 'MDD_4_W',\n",
       " 'MDD_5_W',\n",
       " 'MDD_6_W',\n",
       " 'MDD_7_W',\n",
       " 'MDD_8_W',\n",
       " 'MDD_9_W',\n",
       " 'MDD_10_W',\n",
       " 'MDD_score_W',\n",
       " 'MDD_bin_W']"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# problems:\n",
    "problem = share_missing[share_missing>70]\n",
    "print(len(problem), \" problems\")\n",
    "list(problem.index)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I see that the variables on tree numbers and spacing can also be dropped, since they have too many missing values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop([ \n",
    "    'cocoa_trees_num',\n",
    "    'cocoa_land_dec_how_5ago',\n",
    "    'cocoa_tree_spacing_trees',\n",
    "    'cocoa_tree_spacing_rows',\n",
    "    'cocoa_trees_most'\n",
    "], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I see from the above that the variables on irrigation source is missing (since no one irrigates cocoa). These can be droped.\n",
    "\n",
    "Check that indeed (almost) no one irrigates cocoa:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "no     1373\n",
       "yes       4\n",
       "Name: cocoa_irrigation_yn, dtype: int64"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['cocoa_irrigation_yn'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(list(data.filter(regex = 'irrigation')), axis = 1, inplace = True) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The list also indicated that Minimum Dietary Diversity (MDD) variables is missing for most observations. This is by construction since these questions were asked only of female respondents. As much it is a potential good indicator, it is skipped for the majority of the sample, so I will drop those"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(list(data.filter(regex='MDD')), axis = 1, inplace = True)\n",
    "data.drop(list(data.filter(regex='nutrition')), axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The list also shows many missing for land leased out. It is likely the value is only filled for thos who did lease. Let's check "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    0.0\n",
       "mean     NaN\n",
       "std      NaN\n",
       "min      NaN\n",
       "25%      NaN\n",
       "50%      NaN\n",
       "75%      NaN\n",
       "max      NaN\n",
       "Name: all_land_leasedout_landunit_ha, dtype: float64"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.loc[data['all_land_leasedout_yn']=='no', 'all_land_leasedout_landunit_ha'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    45.000000\n",
       "mean      2.338280\n",
       "std       2.298098\n",
       "min       0.000000\n",
       "25%       1.000000\n",
       "50%       2.000000\n",
       "75%       2.500000\n",
       "max      10.000000\n",
       "Name: all_land_leasedout_landunit_ha, dtype: float64"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.loc[data['all_land_leasedout_yn']=='yes', 'all_land_leasedout_landunit_ha'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's impose `0` if not leased out and `no` if `0` was leased out:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[data['all_land_leasedout_yn']=='no', 'all_land_leasedout_landunit_ha']=0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[data['all_land_leasedout_landunit_ha']==0, 'all_land_leasedout_yn']='no'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The list above also showed missing values for times, hh_hi and gender in manure application. This is likely because the application of manure was `no` for most farmers. Let's double check:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "no     1305\n",
       "yes      72\n",
       "Name: cocoa_manurecomp_yn, dtype: int64"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['cocoa_manurecomp_yn'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "yes    66\n",
       "no      6\n",
       "Name: cocoa_manurecomp_gender_1, dtype: int64"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['cocoa_manurecomp_gender_1'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This is indeed the case. This is actually the case for many `yes/no` indicators, where the detailed question are missing in case the first one is `no`. Therefore we will impose `no` to all missing features related to `no` answer from a previous  `yes/no` question"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['all_land_leasedout_yn',\n",
       " 'cocoa_intercrop_yn',\n",
       " 'cocoa_trees_rows_yn',\n",
       " 'cocoa_landclear_yn',\n",
       " 'cocoa_landprep_yn',\n",
       " 'cocoa_planting_yn',\n",
       " 'cocoa_gfert_yn',\n",
       " 'cocoa_lfert_yn',\n",
       " 'cocoa_manurecomp_yn',\n",
       " 'cocoa_herb_yn',\n",
       " 'cocoa_pest_yn',\n",
       " 'cocoa_fung_yn',\n",
       " 'cocoa_weed_yn',\n",
       " 'cocoa_pruning_yn',\n",
       " 'cocoa_harvest_yn',\n",
       " 'cocoa_podbreak_yn',\n",
       " 'cocoa_ferment_yn',\n",
       " 'cocoa_transport_yn',\n",
       " 'cocoa_dry_yn',\n",
       " 'cocoa_losses_disease_yn',\n",
       " 'cocoa_losses_pests_yn',\n",
       " 'cocoa_losses_storage_yn',\n",
       " 'cocoa_records_yn',\n",
       " 'cocoa_training_5years_yn',\n",
       " 'cocoa_training_lastyear_yn',\n",
       " 'cocoa_certification_clean_yn',\n",
       " 'cocoa_credit_yn',\n",
       " 'loan_lastyear_yn']"
      ]
     },
     "execution_count": 85,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yes_no_questions = list(data.filter(regex = \"_yn\"))\n",
    "yes_no_questions"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's create a supporting function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [],
   "source": [
    "def impose_yn_logic(data, data_properties, main_feature, features_to_clean_main_stub = '', yes_no_label = 'lblYES_NO', verbose = 1):\n",
    "    '''\n",
    "    Imposes yes/no logic in the data. If a main question is `no` all children should also be `no`\n",
    "        \n",
    "    INPUT:\n",
    "        data: panda DataFrame\n",
    "        data_properties: StataRead dictionary object created with pd.read_stata(filename, iterator = True)\n",
    "        main_feature: yes/no parent feature\n",
    "        features_to_clean_main_stub (optional): main stub to find the children features. If not provided the main stub is \n",
    "            derived from the main_feature, removing `_yn` string\n",
    "        yes_no_label: (stata) label that maps YES/NO questions\n",
    "        verbose: if greater than 0, prints informative messages\n",
    "        \n",
    "    OUTPUT:\n",
    "        data: modified panda DataFrame\n",
    "    '''\n",
    "    \n",
    "        \n",
    "    #if no main stub is passed, find the main stub, excluding the \"yn\"\n",
    "    if features_to_clean_main_stub == '':\n",
    "        main_stub = main_feature.replace(\"_yn\",\"\")\n",
    "    else:\n",
    "        main_stub = features_to_clean_main_stub\n",
    "    \n",
    "    # get a list of all other variable with the same stub\n",
    "    features_to_clean = list(data.filter(regex = main_stub))\n",
    "    try:\n",
    "        features_to_clean.remove(main_feature)\n",
    "    except:\n",
    "        ;\n",
    "    \n",
    "    if verbose >=1:\n",
    "        print('')\n",
    "        print(\"Based on {}, clean:\".format(main_feature))\n",
    "        print(features_to_clean)\n",
    "    \n",
    "    \n",
    "    # for each feature in the list, if the feature is a yes/no question, impose 'no' if main is 'no'\n",
    "    for this_feature, this_label in zip(data_properties.varlist, data_properties.lbllist):\n",
    "        if (this_feature in features_to_clean) & (this_label == yes_no_label):\n",
    "            data.loc[data[main_feature]=='no', this_feature] = 'no'\n",
    "            \n",
    "    return data\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Based on all_land_leasedout_yn, clean:\n",
      "['all_land_leasedout_landunit_ha']\n",
      "\n",
      "Based on cocoa_intercrop_yn, clean:\n",
      "['cocoa_intercrop_5yo', 'cocoa_intercrop_5_25yo', 'cocoa_intercrop_25yo', 'cocoa_intercrop_othercrop_1', 'cocoa_intercrop_othercrop_2', 'cocoa_intercrop_othercrop_3', 'cocoa_intercrop_othercrop_4', 'cocoa_intercrop_othercrop_5', 'cocoa_intercrop_othercrop_6', 'cocoa_intercrop_othercrop_7', 'cocoa_intercrop_othercrop_8', 'cocoa_intercrop_othercrop_9', 'cocoa_intercrop_othercrop_10', 'cocoa_intercrop_othercrop_11', 'cocoa_intercrop_othercrop_12', 'cocoa_intercrop_othercrop_13', 'cocoa_intercrop_othercrop_14', 'cocoa_intercrop_othercrop_15', 'cocoa_intercrop_othercrop_16', 'cocoa_intercrop_othercrop_17', 'cocoa_intercrop_othercrop_18', 'cocoa_intercrop_othercrop_19', 'cocoa_intercrop_othercrop_20', 'cocoa_intercrop_othercrop_21', 'cocoa_intercrop_othercrop_22', 'cocoa_intercrop_othercrop_23', 'cocoa_intercrop_othercrop_24', 'cocoa_intercrop_othercrop_25', 'cocoa_intercrop_othercrop_26', 'cocoa_intercrop_othercrop_27', 'cocoa_intercrop_othercrop_28', 'cocoa_intercrop_othercrop_29', 'cocoa_intercrop_othercrop_30', 'cocoa_intercrop_othercrop_31', 'cocoa_intercrop_othercrop_32', 'cocoa_intercrop_othercrop_33', 'cocoa_intercrop_othercrop_34', 'cocoa_intercrop_othercrop_35', 'cocoa_intercrop_othercrop_36', 'cocoa_intercrop_othercrop_37', 'cocoa_intercrop_othercrop_38', 'cocoa_intercrop_othercrop_39', 'cocoa_intercrop_othercrop_40', 'cocoa_intercrop_othercrop_41', 'cocoa_intercrop_othercrop_42', 'cocoa_intercrop_othercrop_43', 'cocoa_intercrop_othercrop_44', 'cocoa_intercrop_othercrop_45', 'cocoa_intercrop_othercrop_46', 'cocoa_intercrop_othercrop_47', 'cocoa_intercrop_othercrop_48', 'cocoa_intercrop_othercrop_49']\n",
      "\n",
      "Based on cocoa_trees_rows_yn, clean:\n",
      "[]\n",
      "\n",
      "Based on cocoa_landclear_yn, clean:\n",
      "['cocoa_landclear_hhhiredcom_1', 'cocoa_landclear_hhhiredcom_2', 'cocoa_landclear_hhhiredcom_3', 'cocoa_landclear_hhhiredcom_4', 'cocoa_landclear_gender_1', 'cocoa_landclear_gender_2']\n",
      "\n",
      "Based on cocoa_landprep_yn, clean:\n",
      "['cocoa_landprep_hhhiredcom_1', 'cocoa_landprep_hhhiredcom_2', 'cocoa_landprep_hhhiredcom_3', 'cocoa_landprep_hhhiredcom_4', 'cocoa_landprep_gender_1', 'cocoa_landprep_gender_2']\n",
      "\n",
      "Based on cocoa_planting_yn, clean:\n",
      "['cocoa_planting_hhhiredcom_1', 'cocoa_planting_hhhiredcom_2', 'cocoa_planting_hhhiredcom_3', 'cocoa_planting_hhhiredcom_4', 'cocoa_planting_gender_1', 'cocoa_planting_gender_2', 'cocoa_planting_seed_origin_1', 'cocoa_planting_seed_origin_2', 'cocoa_planting_seed_origin_3', 'cocoa_planting_seed_origin_4', 'cocoa_planting_seed_origin_5', 'cocoa_planting_seed_origin_6', 'cocoa_planting_seed_origin_7', 'cocoa_planting_seed_origin_8', 'cocoa_planting_seed_origin_9', 'cocoa_planting_seed_origin_10', 'cocoa_planting_seed_origin_999']\n",
      "\n",
      "Based on cocoa_gfert_yn, clean:\n",
      "['cocoa_gfert_hhhiredcom_1', 'cocoa_gfert_hhhiredcom_2', 'cocoa_gfert_hhhiredcom_3', 'cocoa_gfert_hhhiredcom_4', 'cocoa_gfert_gender_1', 'cocoa_gfert_gender_2']\n",
      "\n",
      "Based on cocoa_lfert_yn, clean:\n",
      "['cocoa_lfert_hhhiredcom_1', 'cocoa_lfert_hhhiredcom_2', 'cocoa_lfert_hhhiredcom_3', 'cocoa_lfert_hhhiredcom_4', 'cocoa_lfert_gender_1', 'cocoa_lfert_gender_2']\n",
      "\n",
      "Based on cocoa_manurecomp_yn, clean:\n",
      "['cocoa_manurecomp_hhhiredcom_1', 'cocoa_manurecomp_hhhiredcom_2', 'cocoa_manurecomp_hhhiredcom_3', 'cocoa_manurecomp_hhhiredcom_4', 'cocoa_manurecomp_gender_1', 'cocoa_manurecomp_gender_2']\n",
      "\n",
      "Based on cocoa_herb_yn, clean:\n",
      "['cocoa_herb_hhhiredcom_1', 'cocoa_herb_hhhiredcom_2', 'cocoa_herb_hhhiredcom_3', 'cocoa_herb_hhhiredcom_4', 'cocoa_herb_gender_1', 'cocoa_herb_gender_2']\n",
      "\n",
      "Based on cocoa_pest_yn, clean:\n",
      "['cocoa_pest_hhhiredcom_1', 'cocoa_pest_hhhiredcom_2', 'cocoa_pest_hhhiredcom_3', 'cocoa_pest_hhhiredcom_4', 'cocoa_pest_gender_1', 'cocoa_pest_gender_2']\n",
      "\n",
      "Based on cocoa_fung_yn, clean:\n",
      "['cocoa_fung_hhhiredcom_1', 'cocoa_fung_hhhiredcom_2', 'cocoa_fung_hhhiredcom_3', 'cocoa_fung_hhhiredcom_4', 'cocoa_fung_gender_1', 'cocoa_fung_gender_2']\n",
      "\n",
      "Based on cocoa_weed_yn, clean:\n",
      "['cocoa_weed_hhhiredcom_1', 'cocoa_weed_hhhiredcom_2', 'cocoa_weed_hhhiredcom_3', 'cocoa_weed_hhhiredcom_4', 'cocoa_weed_gender_1', 'cocoa_weed_gender_2']\n",
      "\n",
      "Based on cocoa_pruning_yn, clean:\n",
      "['cocoa_pruning_hhhiredcom_1', 'cocoa_pruning_hhhiredcom_2', 'cocoa_pruning_hhhiredcom_3', 'cocoa_pruning_hhhiredcom_4', 'cocoa_pruning_gender_1', 'cocoa_pruning_gender_2']\n",
      "\n",
      "Based on cocoa_harvest_yn, clean:\n",
      "['cocoa_harvest_months_all_1', 'cocoa_harvest_months_all_2', 'cocoa_harvest_months_all_3', 'cocoa_harvest_months_all_4', 'cocoa_harvest_months_all_5', 'cocoa_harvest_months_all_6', 'cocoa_harvest_months_all_7', 'cocoa_harvest_months_all_8', 'cocoa_harvest_months_all_9', 'cocoa_harvest_months_all_10', 'cocoa_harvest_months_all_11', 'cocoa_harvest_months_all_12', 'cocoa_harvest_months_all_999', 'cocoa_harvest_months_all_99', 'cocoa_harvest_hhhiredcom_1', 'cocoa_harvest_hhhiredcom_2', 'cocoa_harvest_hhhiredcom_3', 'cocoa_harvest_hhhiredcom_4', 'cocoa_harvest_gender_1', 'cocoa_harvest_gender_2']\n",
      "\n",
      "Based on cocoa_podbreak_yn, clean:\n",
      "['cocoa_podbreak_hhhiredcom_1', 'cocoa_podbreak_hhhiredcom_2', 'cocoa_podbreak_hhhiredcom_3', 'cocoa_podbreak_hhhiredcom_4', 'cocoa_podbreak_gender_1', 'cocoa_podbreak_gender_2']\n",
      "\n",
      "Based on cocoa_ferment_yn, clean:\n",
      "['cocoa_ferment_hhhiredcom_1', 'cocoa_ferment_hhhiredcom_2', 'cocoa_ferment_hhhiredcom_3', 'cocoa_ferment_hhhiredcom_4', 'cocoa_ferment_gender_1', 'cocoa_ferment_gender_2']\n",
      "\n",
      "Based on cocoa_transport_yn, clean:\n",
      "['cocoa_transport_hhhiredcom_1', 'cocoa_transport_hhhiredcom_2', 'cocoa_transport_hhhiredcom_3', 'cocoa_transport_hhhiredcom_4', 'cocoa_transport_gender_1', 'cocoa_transport_gender_2']\n",
      "\n",
      "Based on cocoa_dry_yn, clean:\n",
      "['cocoa_dry_hhhiredcom_1', 'cocoa_dry_hhhiredcom_2', 'cocoa_dry_hhhiredcom_3', 'cocoa_dry_hhhiredcom_4', 'cocoa_dry_gender_1', 'cocoa_dry_gender_2']\n",
      "\n",
      "Based on cocoa_losses_disease_yn, clean:\n",
      "['cocoa_losses_disease_type_1', 'cocoa_losses_disease_type_2', 'cocoa_losses_disease_type_3', 'cocoa_losses_disease_type_5', 'cocoa_losses_disease_type_4', 'cocoa_losses_disease_affect']\n",
      "\n",
      "Based on cocoa_losses_pests_yn, clean:\n",
      "['cocoa_losses_pests_type_1', 'cocoa_losses_pests_type_2', 'cocoa_losses_pests_type_3', 'cocoa_losses_pests_type_4', 'cocoa_losses_pests_type_5', 'cocoa_losses_pests_type_6', 'cocoa_losses_pests_type_7', 'cocoa_losses_pests_type_9', 'cocoa_losses_pests_type_10', 'cocoa_losses_pests_type_999', 'cocoa_losses_pests_affect']\n",
      "\n",
      "Based on cocoa_losses_storage_yn, clean:\n",
      "['cocoa_losses_storage_affect']\n",
      "\n",
      "Based on cocoa_records_yn, clean:\n",
      "['cocoa_records_type_1', 'cocoa_records_type_2', 'cocoa_records_type_3', 'cocoa_records_type_4', 'cocoa_records_type_5', 'cocoa_records_type_999']\n",
      "\n",
      "Based on cocoa_training_5years_yn, clean:\n",
      "[]\n",
      "\n",
      "Based on cocoa_training_lastyear_yn, clean:\n",
      "[]\n",
      "\n",
      "Based on cocoa_certification_clean_yn, clean:\n",
      "[]\n",
      "\n",
      "Based on cocoa_credit_yn, clean:\n",
      "['cocoa_credit_inputs_1', 'cocoa_credit_inputs_2', 'cocoa_credit_inputs_3', 'cocoa_credit_inputs_4', 'cocoa_credit_whom']\n",
      "\n",
      "Based on loan_lastyear_yn, clean:\n",
      "[]\n"
     ]
    }
   ],
   "source": [
    "for main_question in yes_no_questions:\n",
    "    \n",
    "    # for each feature in the list, if the feature is a yes/no question, impose 'no' if main is 'no'\n",
    "    data = impose_yn_logic(data, data_properties, main_question)\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I see that `cocoa_credit_whom` is an specification from `cocoa_credit_yn`, but it is not a yes/no question. Since it only provides some additional infomation for those taking inputs on credit, I will drop it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(list(data.filter(regex = 'cocoa_credit_whom')), axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I also see from above that features `cocoa_trees_rows_yn`, `cocoa_training_5years_yn`, `cocoa_training_lastyear_yn`, `cocoa_certification_clean_yn`, `loan_lastyear_yn` didn't trigger any additional adjustments. Except for `cocoa_trees_rows_yn` and `cocoa_training_lastyear_yn` (no dependents) these are special cases, where feature naming convention is different. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Based on loan_lastyear_yn, clean:\n",
      "['loan_source_1', 'loan_source_2', 'loan_source_3', 'loan_source_4', 'loan_source_5', 'loan_source_6', 'loan_source_7', 'loan_source_8', 'loan_source_9', 'loan_source_10', 'loan_source_11', 'loan_source_12', 'loan_source_13', 'loan_source_999', 'loan_amount_usd', 'loan_repayment_success', 'loan_repayment_c', 'loan_use_1', 'loan_use_2', 'loan_use_3', 'loan_use_4', 'loan_use_5', 'loan_use_6', 'loan_use_7', 'loan_use_8', 'loan_use_9', 'loan_use_10', 'loan_use_11', 'loan_use_12', 'loan_use_13', 'loan_use_14', 'loan_use_15', 'loan_use_16', 'loan_use_17', 'loan_use_19', 'loan_use_20', 'loan_use_18', 'loan_use_999', 'noloan_whynot_1', 'noloan_whynot_2', 'noloan_whynot_3', 'noloan_whynot_4', 'noloan_whynot_5', 'noloan_whynot_6', 'noloan_whynot_7', 'noloan_whynot_8']\n",
      "\n",
      "Based on cocoa_training_5years_yn, clean:\n",
      "['cocoa_training_days_cat', 'cocoa_training_lastyear_yn', 'cocoa_training_topic_1', 'cocoa_training_topic_2', 'cocoa_training_topic_3', 'cocoa_training_topic_4', 'cocoa_training_topic_5', 'cocoa_training_topic_6', 'cocoa_training_topic_7', 'cocoa_training_topic_8', 'cocoa_training_topic_9', 'cocoa_training_topic_10', 'cocoa_training_topic_11', 'cocoa_training_source_1', 'cocoa_training_source_2', 'cocoa_training_source_3', 'cocoa_training_source_4', 'cocoa_training_source_5', 'cocoa_training_source_6', 'cocoa_training_source_999']\n",
      "\n",
      "Based on cocoa_certification_clean_yn, clean:\n",
      "['cocoa_cert_type_clean_1', 'cocoa_cert_type_clean_2', 'cocoa_cert_type_clean_3', 'cocoa_cert_type_clean_4']\n"
     ]
    }
   ],
   "source": [
    "data = impose_yn_logic(data, data_properties, 'loan_lastyear_yn',  'loan')\n",
    "data = impose_yn_logic(data, data_properties, 'cocoa_training_5years_yn',  'cocoa_training_')\n",
    "data = impose_yn_logic(data, data_properties, 'cocoa_certification_clean_yn',  'cocoa_cert_type_clean_')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I can see from above that there are features related to `noloan`. Since this is undefined for those who did take a loan, I will drop it to keep only features which are meaningfull for the whole sample. Similar argument applies to loan collateral and loan repayment success"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(list(data.filter(regex = 'noloan')), axis = 1, inplace = True)\n",
    "data.drop(list(data.filter(regex = 'loan_repayment')), axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I can also see that `loan_amount_` is there, and this should be 0 if no loan was taken"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[data['loan_lastyear_yn']=='no', 'loan_amount_usd']=0\n",
    "data.loc[data['loan_amount_usd']==0, 'loan_lastyear_yn']='no'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly I see that `cocoa_training_days_cat` should be 0 if no training was taken:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[data['cocoa_training_5years_yn']=='no', 'cocoa_training_days_cat']='none'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I had also seen that the features on `cocoa_prodgroup_services_` are missing. These are specification of services provided by producer groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Based on cocoa_prodgroup_hhmem, clean:\n",
      "['cocoa_prodgroup_services_1', 'cocoa_prodgroup_services_2', 'cocoa_prodgroup_services_3', 'cocoa_prodgroup_services_4', 'cocoa_prodgroup_services_5', 'cocoa_prodgroup_services_6', 'cocoa_prodgroup_services_7', 'cocoa_prodgroup_services_8', 'cocoa_prodgroup_services_9', 'cocoa_prodgroup_services_10', 'cocoa_prodgroup_services_999']\n"
     ]
    }
   ],
   "source": [
    "data = impose_yn_logic(data, data_properties, 'cocoa_prodgroup_hhmem',  'cocoa_prodgroup_services_')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's check what is over:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count    630.000000\n",
      "mean       1.556178\n",
      "std        8.413180\n",
      "min        0.000000\n",
      "25%        0.000000\n",
      "50%        0.000000\n",
      "75%        0.000000\n",
      "90%        1.161946\n",
      "95%        3.848947\n",
      "99%       55.908497\n",
      "max       86.347131\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    " # check missing per variables:\n",
    "share_missing = missing_per_variable(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_missing(share_missing)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We see that `_affect` variables are missing. They measure how affected farmer were by different types of losses. Since this is only relevant when they occur, let's drop them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(list(data.filter(regex = '_affect')), axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We also see that a few of the `hh_` variables are missing. These are variables that were only collected in Côte d'Ivoire"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 97,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(['hh_fan', 'hh_tractor', 'hh_gaselectric_stove', 'hh_cable_subscription'],\n",
    "         axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Similarly to `PPI_190` and `PPI_310`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(['PPI_190dollar_day', 'PPI_310dollar_day'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Actually I will drop all `DHS` and `PPI` indices variables, since they are asset-based indices, and I want to predict achieving the benchmark regardless of those"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 99,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(list(data.filter(regex = 'dhs')), axis = 1, inplace = True)\n",
    "data.drop(list(data.filter(regex = 'PPI')), axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some detailed features on `mosquito nets` are also dependent on having said `yes` to mosquito net. Let's impose the logic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 100,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Based on hh_mosquito_net, clean:\n",
      "['hh_mosquito_net_num', 'hh_mosquito_net_lastnight']\n"
     ]
    }
   ],
   "source": [
    "data = impose_yn_logic(data, data_properties, 'hh_mosquito_net')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[data['hh_mosquito_net']=='no', 'hh_mosquito_net_num']=0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The feature on `cocoa_intercrop_25yo` seems to have too many missing, even after the logic imposing. I will drop it the whole set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 102,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(['cocoa_intercrop_5yo', 'cocoa_intercrop_5_25yo', 'cocoa_intercrop_25yo'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The feature `cocoa_land_inc_how_5ago` applies only to farmers that increased land. So I will drop to keep the features generic"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 103,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.drop(['cocoa_land_inc_how_5ago'], axis = 1, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's impose that land was not leased out if missing, since this was the most common answer:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 104,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[data['all_land_leasedout_yn'].isnull(), 'all_land_leasedout_yn']='no'\n",
    "data.loc[data['all_land_leasedout_yn']=='no', 'all_land_leasedout_landunit_ha']=0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The indicator of all children at school only applies to farmers with children in school age. To make it complete, I will set all children at school to `yes` if no childrens present, since this is the more neutral answer in this context"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 105,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[data['male_children_5_14']==0, 'male_children_5_14_school']='yes'\n",
    "data.loc[data['female_children_5_14']==0, 'female_children_5_14_school']='yes'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "I'll set none for `crops_important2` where missing:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 106,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.loc[data['crops_important2'].isnull(), 'crops_important2']='none/ not applicable'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's do a final check:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 107,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "count    609.000000\n",
      "mean       0.267233\n",
      "std        1.019338\n",
      "min        0.000000\n",
      "25%        0.000000\n",
      "50%        0.000000\n",
      "75%        0.000000\n",
      "90%        0.871460\n",
      "95%        2.251271\n",
      "99%        3.921569\n",
      "max       15.177923\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    " # check missing per variables:\n",
    "share_missing = missing_per_variable(data)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 108,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plot_missing(share_missing)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 4 Final Details"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Before finishing, I'll replace the binary categories by their numeric values (0 or 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 109,
   "metadata": {},
   "outputs": [],
   "source": [
    "def convert_tocodes(data, data_properties, label_name):\n",
    "    '''\n",
    "    Convert categories to numeric values, according to the stata dictionaty\n",
    "        \n",
    "    INPUT:\n",
    "        data: panda DataFrame\n",
    "        data_properties: StataRead dictionary object created with pd.read_stata(filename, iterator = True)\n",
    "        label_name: (stata) label that maps categories in that question\n",
    "        \n",
    "    OUTPUT:\n",
    "        data: modified panda DataFrame\n",
    "    '''\n",
    "    \n",
    "    # make a dataframe\n",
    "    data_and_label = pd.DataFrame({\n",
    "        \"variable label\": data_properties.varlist,\n",
    "        \"value label\": data_properties.lbllist})\n",
    "    \n",
    "    # filter on the selected label\n",
    "    data_and_label = data_and_label[data_and_label['value label']==label_name]\n",
    "\n",
    "    # convert all features that use that label\n",
    "    for this_column in list(data.columns):\n",
    "\n",
    "        if this_column in list(data_and_label['variable label']):\n",
    "            data[this_column] = data[this_column].cat.codes\n",
    "            data.loc[data[this_column] == -1, this_column] = np.nan\n",
    "            \n",
    "    return data\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see how they look like before applying the function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "male      1227\n",
       "female     150\n",
       "Name: head_gender, dtype: int64"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['head_gender'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "no     926\n",
       "yes    451\n",
       "Name: cocoa_gfert_yn, dtype: int64"
      ]
     },
     "execution_count": 111,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['cocoa_gfert_yn'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "data = convert_tocodes(data, data_properties, \"lblYES_NO\")\n",
    "data = convert_tocodes(data, data_properties, \"lblMALE_FEMALE\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's see how they look like AFTER applying the function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    1227\n",
       "1.0     150\n",
       "Name: head_gender, dtype: int64"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['head_gender'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0    926\n",
       "1.0    451\n",
       "Name: cocoa_gfert_yn, dtype: int64"
      ]
     },
     "execution_count": 115,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['cocoa_gfert_yn'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's rename the columns to a more meaningful values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['country',\n",
       " 'region',\n",
       " 'hhmem_number',\n",
       " 'male_adults_over_65',\n",
       " 'female_adults_over_65',\n",
       " 'male_adults_15_65',\n",
       " 'female_adults_15_65',\n",
       " 'male_children_5_14',\n",
       " 'female_children_5_14',\n",
       " 'male_children_0_4',\n",
       " 'female_children_0_4',\n",
       " 'dep_ratio',\n",
       " 'head_age',\n",
       " 'head_age_cat',\n",
       " 'head_gender',\n",
       " 'head_education',\n",
       " 'male_children_5_14_school',\n",
       " 'female_children_5_14_school',\n",
       " 'incsource_biggest1',\n",
       " 'incsource_biggest2',\n",
       " 'hh_income_perc_cocoa',\n",
       " 'hh_income_perc_othercrops',\n",
       " 'hh_income_perc_livestock',\n",
       " 'hh_income_perc_fish',\n",
       " 'hh_income_perc_bushproducts',\n",
       " 'hh_income_perc_farmlabour',\n",
       " 'hh_income_perc_nonaglabour',\n",
       " 'hh_income_perc_smallbusiness',\n",
       " 'hh_income_perc_comp',\n",
       " 'hh_income_perc_govt',\n",
       " 'hh_income_perc_landsalelease',\n",
       " 'hh_income_perc_remittances',\n",
       " 'hh_income_perc_other',\n",
       " 'crops_all_produced_1',\n",
       " 'crops_all_produced_2',\n",
       " 'crops_all_produced_3',\n",
       " 'crops_all_produced_4',\n",
       " 'crops_all_produced_5',\n",
       " 'crops_all_produced_6',\n",
       " 'crops_all_produced_7',\n",
       " 'crops_all_produced_8',\n",
       " 'crops_all_produced_9',\n",
       " 'crops_all_produced_10',\n",
       " 'crops_all_produced_11',\n",
       " 'crops_all_produced_12',\n",
       " 'crops_all_produced_13',\n",
       " 'crops_all_produced_14',\n",
       " 'crops_all_produced_15',\n",
       " 'crops_all_produced_16',\n",
       " 'crops_all_produced_17',\n",
       " 'crops_all_produced_18',\n",
       " 'crops_all_produced_19',\n",
       " 'crops_all_produced_20',\n",
       " 'crops_all_produced_21',\n",
       " 'crops_all_produced_22',\n",
       " 'crops_all_produced_23',\n",
       " 'crops_all_produced_24',\n",
       " 'crops_all_produced_25',\n",
       " 'crops_all_produced_26',\n",
       " 'crops_all_produced_27',\n",
       " 'crops_all_produced_28',\n",
       " 'crops_all_produced_29',\n",
       " 'crops_all_produced_30',\n",
       " 'crops_all_produced_31',\n",
       " 'crops_all_produced_32',\n",
       " 'crops_all_produced_33',\n",
       " 'crops_all_produced_34',\n",
       " 'crops_all_produced_35',\n",
       " 'crops_all_produced_36',\n",
       " 'crops_all_produced_37',\n",
       " 'crops_all_produced_38',\n",
       " 'crops_all_produced_39',\n",
       " 'crops_all_produced_40',\n",
       " 'crops_all_produced_41',\n",
       " 'crops_all_produced_42',\n",
       " 'crops_all_produced_43',\n",
       " 'crops_all_produced_44',\n",
       " 'crops_all_produced_45',\n",
       " 'crops_all_produced_46',\n",
       " 'crops_all_produced_47',\n",
       " 'crops_all_produced_48',\n",
       " 'crops_all_produced_49',\n",
       " 'n_crops_produced',\n",
       " 'crops_all_sold_1',\n",
       " 'crops_all_sold_2',\n",
       " 'crops_all_sold_3',\n",
       " 'crops_all_sold_4',\n",
       " 'crops_all_sold_5',\n",
       " 'crops_all_sold_6',\n",
       " 'crops_all_sold_7',\n",
       " 'crops_all_sold_8',\n",
       " 'crops_all_sold_9',\n",
       " 'crops_all_sold_10',\n",
       " 'crops_all_sold_11',\n",
       " 'crops_all_sold_12',\n",
       " 'crops_all_sold_13',\n",
       " 'crops_all_sold_14',\n",
       " 'crops_all_sold_15',\n",
       " 'crops_all_sold_16',\n",
       " 'crops_all_sold_17',\n",
       " 'crops_all_sold_18',\n",
       " 'crops_all_sold_19',\n",
       " 'crops_all_sold_20',\n",
       " 'crops_all_sold_21',\n",
       " 'crops_all_sold_22',\n",
       " 'crops_all_sold_23',\n",
       " 'crops_all_sold_24',\n",
       " 'crops_all_sold_25',\n",
       " 'crops_all_sold_26',\n",
       " 'crops_all_sold_27',\n",
       " 'crops_all_sold_28',\n",
       " 'crops_all_sold_29',\n",
       " 'crops_all_sold_30',\n",
       " 'crops_all_sold_31',\n",
       " 'crops_all_sold_32',\n",
       " 'crops_all_sold_33',\n",
       " 'crops_all_sold_34',\n",
       " 'crops_all_sold_35',\n",
       " 'crops_all_sold_36',\n",
       " 'crops_all_sold_37',\n",
       " 'crops_all_sold_38',\n",
       " 'crops_all_sold_39',\n",
       " 'crops_all_sold_40',\n",
       " 'crops_all_sold_41',\n",
       " 'crops_all_sold_42',\n",
       " 'crops_all_sold_43',\n",
       " 'crops_all_sold_44',\n",
       " 'crops_all_sold_45',\n",
       " 'crops_all_sold_46',\n",
       " 'crops_all_sold_47',\n",
       " 'crops_all_sold_48',\n",
       " 'crops_all_sold_49',\n",
       " 'n_crops_sold',\n",
       " 'crops_important2',\n",
       " 'all_land_cultivated_ha',\n",
       " 'all_land_owned_ha',\n",
       " 'all_land_leased_ha',\n",
       " 'all_land_scrop_abunu_ha',\n",
       " 'all_land_scrop_abusa_ha',\n",
       " 'all_land_fallow_ha',\n",
       " 'all_land_numlandparcels',\n",
       " 'all_land_leasedout_yn',\n",
       " 'all_land_leasedout_landunit_ha',\n",
       " 'cocoa_land_used_ha',\n",
       " 'cocoa_land_used_morethan5_ha',\n",
       " 'cocoa_land_owned_ha',\n",
       " 'cocoa_land_leased_ha',\n",
       " 'cocoa_land_scrop_abunu_ha',\n",
       " 'cocoa_land_scrop_abusa_ha',\n",
       " 'cocoa_land_numlandparcels',\n",
       " 'cocoa_years',\n",
       " 'cocoa_trees_age_5yo_ha',\n",
       " 'cocoa_trees_age_5_25yo_ha',\n",
       " 'cocoa_trees_age_25yo_ha',\n",
       " 'cocoa_land_incdecnc_5ago',\n",
       " 'cocoa_intercrop_yn',\n",
       " 'cocoa_intercrop_othercrop_1',\n",
       " 'cocoa_intercrop_othercrop_2',\n",
       " 'cocoa_intercrop_othercrop_3',\n",
       " 'cocoa_intercrop_othercrop_4',\n",
       " 'cocoa_intercrop_othercrop_5',\n",
       " 'cocoa_intercrop_othercrop_6',\n",
       " 'cocoa_intercrop_othercrop_7',\n",
       " 'cocoa_intercrop_othercrop_8',\n",
       " 'cocoa_intercrop_othercrop_9',\n",
       " 'cocoa_intercrop_othercrop_10',\n",
       " 'cocoa_intercrop_othercrop_11',\n",
       " 'cocoa_intercrop_othercrop_12',\n",
       " 'cocoa_intercrop_othercrop_13',\n",
       " 'cocoa_intercrop_othercrop_14',\n",
       " 'cocoa_intercrop_othercrop_15',\n",
       " 'cocoa_intercrop_othercrop_16',\n",
       " 'cocoa_intercrop_othercrop_17',\n",
       " 'cocoa_intercrop_othercrop_18',\n",
       " 'cocoa_intercrop_othercrop_19',\n",
       " 'cocoa_intercrop_othercrop_20',\n",
       " 'cocoa_intercrop_othercrop_21',\n",
       " 'cocoa_intercrop_othercrop_22',\n",
       " 'cocoa_intercrop_othercrop_23',\n",
       " 'cocoa_intercrop_othercrop_24',\n",
       " 'cocoa_intercrop_othercrop_25',\n",
       " 'cocoa_intercrop_othercrop_26',\n",
       " 'cocoa_intercrop_othercrop_27',\n",
       " 'cocoa_intercrop_othercrop_28',\n",
       " 'cocoa_intercrop_othercrop_29',\n",
       " 'cocoa_intercrop_othercrop_30',\n",
       " 'cocoa_intercrop_othercrop_31',\n",
       " 'cocoa_intercrop_othercrop_32',\n",
       " 'cocoa_intercrop_othercrop_33',\n",
       " 'cocoa_intercrop_othercrop_34',\n",
       " 'cocoa_intercrop_othercrop_35',\n",
       " 'cocoa_intercrop_othercrop_36',\n",
       " 'cocoa_intercrop_othercrop_37',\n",
       " 'cocoa_intercrop_othercrop_38',\n",
       " 'cocoa_intercrop_othercrop_39',\n",
       " 'cocoa_intercrop_othercrop_40',\n",
       " 'cocoa_intercrop_othercrop_41',\n",
       " 'cocoa_intercrop_othercrop_42',\n",
       " 'cocoa_intercrop_othercrop_43',\n",
       " 'cocoa_intercrop_othercrop_44',\n",
       " 'cocoa_intercrop_othercrop_45',\n",
       " 'cocoa_intercrop_othercrop_46',\n",
       " 'cocoa_intercrop_othercrop_47',\n",
       " 'cocoa_intercrop_othercrop_48',\n",
       " 'cocoa_intercrop_othercrop_49',\n",
       " 'cocoa_prod_total_kgs',\n",
       " 'cocoa_prod_total_kgsha',\n",
       " 'cocoa_varieties_1',\n",
       " 'cocoa_varieties_2',\n",
       " 'cocoa_varieties_3',\n",
       " 'cocoa_varieties_7',\n",
       " 'cocoa_varieties_4',\n",
       " 'cocoa_varieties_5',\n",
       " 'cocoa_varieties_6',\n",
       " 'cocoa_varieties_999',\n",
       " 'cocoa_trees_rows_yn',\n",
       " 'cocoa_landclear_yn',\n",
       " 'cocoa_landclear_hhhiredcom_1',\n",
       " 'cocoa_landclear_hhhiredcom_2',\n",
       " 'cocoa_landclear_hhhiredcom_3',\n",
       " 'cocoa_landclear_hhhiredcom_4',\n",
       " 'cocoa_landclear_gender_1',\n",
       " 'cocoa_landclear_gender_2',\n",
       " 'cocoa_landprep_yn',\n",
       " 'cocoa_landprep_hhhiredcom_1',\n",
       " 'cocoa_landprep_hhhiredcom_2',\n",
       " 'cocoa_landprep_hhhiredcom_3',\n",
       " 'cocoa_landprep_hhhiredcom_4',\n",
       " 'cocoa_landprep_gender_1',\n",
       " 'cocoa_landprep_gender_2',\n",
       " 'cocoa_planting_yn',\n",
       " 'cocoa_planting_hhhiredcom_1',\n",
       " 'cocoa_planting_hhhiredcom_2',\n",
       " 'cocoa_planting_hhhiredcom_3',\n",
       " 'cocoa_planting_hhhiredcom_4',\n",
       " 'cocoa_planting_gender_1',\n",
       " 'cocoa_planting_gender_2',\n",
       " 'cocoa_planting_seed_origin_1',\n",
       " 'cocoa_planting_seed_origin_2',\n",
       " 'cocoa_planting_seed_origin_3',\n",
       " 'cocoa_planting_seed_origin_4',\n",
       " 'cocoa_planting_seed_origin_5',\n",
       " 'cocoa_planting_seed_origin_6',\n",
       " 'cocoa_planting_seed_origin_7',\n",
       " 'cocoa_planting_seed_origin_8',\n",
       " 'cocoa_planting_seed_origin_9',\n",
       " 'cocoa_planting_seed_origin_10',\n",
       " 'cocoa_planting_seed_origin_999',\n",
       " 'cocoa_gfert_yn',\n",
       " 'cocoa_gfert_hhhiredcom_1',\n",
       " 'cocoa_gfert_hhhiredcom_2',\n",
       " 'cocoa_gfert_hhhiredcom_3',\n",
       " 'cocoa_gfert_hhhiredcom_4',\n",
       " 'cocoa_gfert_gender_1',\n",
       " 'cocoa_gfert_gender_2',\n",
       " 'cocoa_lfert_yn',\n",
       " 'cocoa_lfert_hhhiredcom_1',\n",
       " 'cocoa_lfert_hhhiredcom_2',\n",
       " 'cocoa_lfert_hhhiredcom_3',\n",
       " 'cocoa_lfert_hhhiredcom_4',\n",
       " 'cocoa_lfert_gender_1',\n",
       " 'cocoa_lfert_gender_2',\n",
       " 'cocoa_manurecomp_yn',\n",
       " 'cocoa_manurecomp_hhhiredcom_1',\n",
       " 'cocoa_manurecomp_hhhiredcom_2',\n",
       " 'cocoa_manurecomp_hhhiredcom_3',\n",
       " 'cocoa_manurecomp_hhhiredcom_4',\n",
       " 'cocoa_manurecomp_gender_1',\n",
       " 'cocoa_manurecomp_gender_2',\n",
       " 'cocoa_herb_yn',\n",
       " 'cocoa_herb_hhhiredcom_1',\n",
       " 'cocoa_herb_hhhiredcom_2',\n",
       " 'cocoa_herb_hhhiredcom_3',\n",
       " 'cocoa_herb_hhhiredcom_4',\n",
       " 'cocoa_herb_gender_1',\n",
       " 'cocoa_herb_gender_2',\n",
       " 'cocoa_pest_yn',\n",
       " 'cocoa_pest_hhhiredcom_1',\n",
       " 'cocoa_pest_hhhiredcom_2',\n",
       " 'cocoa_pest_hhhiredcom_3',\n",
       " 'cocoa_pest_hhhiredcom_4',\n",
       " 'cocoa_pest_gender_1',\n",
       " 'cocoa_pest_gender_2',\n",
       " 'cocoa_fung_yn',\n",
       " 'cocoa_fung_hhhiredcom_1',\n",
       " 'cocoa_fung_hhhiredcom_2',\n",
       " 'cocoa_fung_hhhiredcom_3',\n",
       " 'cocoa_fung_hhhiredcom_4',\n",
       " 'cocoa_fung_gender_1',\n",
       " 'cocoa_fung_gender_2',\n",
       " 'cocoa_weed_yn',\n",
       " 'cocoa_weed_hhhiredcom_1',\n",
       " 'cocoa_weed_hhhiredcom_2',\n",
       " 'cocoa_weed_hhhiredcom_3',\n",
       " 'cocoa_weed_hhhiredcom_4',\n",
       " 'cocoa_weed_gender_1',\n",
       " 'cocoa_weed_gender_2',\n",
       " 'cocoa_pruning_yn',\n",
       " 'cocoa_pruning_hhhiredcom_1',\n",
       " 'cocoa_pruning_hhhiredcom_2',\n",
       " 'cocoa_pruning_hhhiredcom_3',\n",
       " 'cocoa_pruning_hhhiredcom_4',\n",
       " 'cocoa_pruning_gender_1',\n",
       " 'cocoa_pruning_gender_2',\n",
       " 'cocoa_harvest_yn',\n",
       " 'cocoa_harvest_months_all_1',\n",
       " 'cocoa_harvest_months_all_2',\n",
       " 'cocoa_harvest_months_all_3',\n",
       " 'cocoa_harvest_months_all_4',\n",
       " 'cocoa_harvest_months_all_5',\n",
       " 'cocoa_harvest_months_all_6',\n",
       " 'cocoa_harvest_months_all_7',\n",
       " 'cocoa_harvest_months_all_8',\n",
       " 'cocoa_harvest_months_all_9',\n",
       " 'cocoa_harvest_months_all_10',\n",
       " 'cocoa_harvest_months_all_11',\n",
       " 'cocoa_harvest_months_all_12',\n",
       " 'cocoa_harvest_months_all_999',\n",
       " 'cocoa_harvest_months_all_99',\n",
       " 'cocoa_harvest_hhhiredcom_1',\n",
       " 'cocoa_harvest_hhhiredcom_2',\n",
       " 'cocoa_harvest_hhhiredcom_3',\n",
       " 'cocoa_harvest_hhhiredcom_4',\n",
       " 'cocoa_harvest_gender_1',\n",
       " 'cocoa_harvest_gender_2',\n",
       " 'cocoa_podbreak_yn',\n",
       " 'cocoa_podbreak_hhhiredcom_1',\n",
       " 'cocoa_podbreak_hhhiredcom_2',\n",
       " 'cocoa_podbreak_hhhiredcom_3',\n",
       " 'cocoa_podbreak_hhhiredcom_4',\n",
       " 'cocoa_podbreak_gender_1',\n",
       " 'cocoa_podbreak_gender_2',\n",
       " 'cocoa_ferment_yn',\n",
       " 'cocoa_ferment_hhhiredcom_1',\n",
       " 'cocoa_ferment_hhhiredcom_2',\n",
       " 'cocoa_ferment_hhhiredcom_3',\n",
       " 'cocoa_ferment_hhhiredcom_4',\n",
       " 'cocoa_ferment_gender_1',\n",
       " 'cocoa_ferment_gender_2',\n",
       " 'cocoa_transport_yn',\n",
       " 'cocoa_transport_hhhiredcom_1',\n",
       " 'cocoa_transport_hhhiredcom_2',\n",
       " 'cocoa_transport_hhhiredcom_3',\n",
       " 'cocoa_transport_hhhiredcom_4',\n",
       " 'cocoa_transport_gender_1',\n",
       " 'cocoa_transport_gender_2',\n",
       " 'cocoa_dry_yn',\n",
       " 'cocoa_dry_hhhiredcom_1',\n",
       " 'cocoa_dry_hhhiredcom_2',\n",
       " 'cocoa_dry_hhhiredcom_3',\n",
       " 'cocoa_dry_hhhiredcom_4',\n",
       " 'cocoa_dry_gender_1',\n",
       " 'cocoa_dry_gender_2',\n",
       " 'cocoa_losses_disease_yn',\n",
       " 'cocoa_losses_pests_yn',\n",
       " 'cocoa_losses_storage_yn',\n",
       " 'cocoa_losses_disease_type_1',\n",
       " 'cocoa_losses_disease_type_2',\n",
       " 'cocoa_losses_disease_type_3',\n",
       " 'cocoa_losses_disease_type_5',\n",
       " 'cocoa_losses_disease_type_4',\n",
       " 'cocoa_losses_pests_type_1',\n",
       " 'cocoa_losses_pests_type_2',\n",
       " 'cocoa_losses_pests_type_3',\n",
       " 'cocoa_losses_pests_type_4',\n",
       " 'cocoa_losses_pests_type_5',\n",
       " 'cocoa_losses_pests_type_6',\n",
       " 'cocoa_losses_pests_type_7',\n",
       " 'cocoa_losses_pests_type_9',\n",
       " 'cocoa_losses_pests_type_10',\n",
       " 'cocoa_losses_pests_type_999',\n",
       " 'cocoa_records_yn',\n",
       " 'cocoa_records_type_1',\n",
       " 'cocoa_records_type_2',\n",
       " 'cocoa_records_type_3',\n",
       " 'cocoa_records_type_4',\n",
       " 'cocoa_records_type_5',\n",
       " 'cocoa_records_type_999',\n",
       " 'cocoa_decisions_gender_1',\n",
       " 'cocoa_decisions_gender_2',\n",
       " 'cocoa_sold_share',\n",
       " 'cocoa_consumed_share',\n",
       " 'cocoa_oblig_share',\n",
       " 'cocoa_givenaway_share',\n",
       " 'cocoa_barterexchange_share',\n",
       " 'cocoa_givenforlabour_share',\n",
       " 'cocoa_lost_share',\n",
       " 'cocoa_price_usdkg',\n",
       " 'cocoa_marketing_months_1',\n",
       " 'cocoa_marketing_months_2',\n",
       " 'cocoa_marketing_months_3',\n",
       " 'cocoa_marketing_months_4',\n",
       " 'cocoa_marketing_months_5',\n",
       " 'cocoa_marketing_months_6',\n",
       " 'cocoa_marketing_months_7',\n",
       " 'cocoa_marketing_months_8',\n",
       " 'cocoa_marketing_months_9',\n",
       " 'cocoa_marketing_months_10',\n",
       " 'cocoa_marketing_months_11',\n",
       " 'cocoa_marketing_months_12',\n",
       " 'cocoa_marketing_months_999',\n",
       " 'cocoa_marketing_months_99',\n",
       " 'cocoa_buyers_1',\n",
       " 'cocoa_buyers_2',\n",
       " 'cocoa_buyers_3',\n",
       " 'cocoa_buyers_4',\n",
       " 'cocoa_buyers_5',\n",
       " 'cocoa_buyers_6',\n",
       " 'cocoa_buyers_7',\n",
       " 'cocoa_buyers_8',\n",
       " 'cocoa_buyers_9',\n",
       " 'cocoa_buyers_10',\n",
       " 'cocoa_buyers_13',\n",
       " 'cocoa_buyers_14',\n",
       " 'cocoa_buyers_15',\n",
       " 'cocoa_buyers_16',\n",
       " 'cocoa_buyers_11',\n",
       " 'cocoa_buyers_12',\n",
       " 'cocoa_buyers_17',\n",
       " 'cocoa_buyers_999',\n",
       " 'cocoa_marketing_gender_1',\n",
       " 'cocoa_marketing_gender_2',\n",
       " 'cocoa_training_5years_yn',\n",
       " 'cocoa_training_days_cat',\n",
       " 'cocoa_training_lastyear_yn',\n",
       " 'cocoa_training_topic_1',\n",
       " 'cocoa_training_topic_2',\n",
       " 'cocoa_training_topic_3',\n",
       " 'cocoa_training_topic_4',\n",
       " 'cocoa_training_topic_5',\n",
       " 'cocoa_training_topic_6',\n",
       " 'cocoa_training_topic_7',\n",
       " 'cocoa_training_topic_8',\n",
       " 'cocoa_training_topic_9',\n",
       " 'cocoa_training_topic_10',\n",
       " 'cocoa_training_topic_11',\n",
       " 'cocoa_training_source_1',\n",
       " 'cocoa_training_source_2',\n",
       " 'cocoa_training_source_3',\n",
       " 'cocoa_training_source_4',\n",
       " 'cocoa_training_source_5',\n",
       " 'cocoa_training_source_6',\n",
       " 'cocoa_training_source_999',\n",
       " 'cocoa_certification_clean_yn',\n",
       " 'cocoa_cert_type_clean_1',\n",
       " 'cocoa_cert_type_clean_2',\n",
       " 'cocoa_cert_type_clean_3',\n",
       " 'cocoa_cert_type_clean_4',\n",
       " 'cocoa_prodgroup_hhmem',\n",
       " 'cocoa_prodgroup_services_1',\n",
       " 'cocoa_prodgroup_services_2',\n",
       " 'cocoa_prodgroup_services_3',\n",
       " 'cocoa_prodgroup_services_4',\n",
       " 'cocoa_prodgroup_services_5',\n",
       " 'cocoa_prodgroup_services_6',\n",
       " 'cocoa_prodgroup_services_7',\n",
       " 'cocoa_prodgroup_services_8',\n",
       " 'cocoa_prodgroup_services_9',\n",
       " 'cocoa_prodgroup_services_10',\n",
       " 'cocoa_prodgroup_services_999',\n",
       " 'cocoa_credit_yn',\n",
       " 'cocoa_credit_inputs_1',\n",
       " 'cocoa_credit_inputs_2',\n",
       " 'cocoa_credit_inputs_3',\n",
       " 'cocoa_credit_inputs_4',\n",
       " 'loan_lastyear_yn',\n",
       " 'loan_source_1',\n",
       " 'loan_source_2',\n",
       " 'loan_source_3',\n",
       " 'loan_source_4',\n",
       " 'loan_source_5',\n",
       " 'loan_source_6',\n",
       " 'loan_source_7',\n",
       " 'loan_source_8',\n",
       " 'loan_source_9',\n",
       " 'loan_source_10',\n",
       " 'loan_source_11',\n",
       " 'loan_source_12',\n",
       " 'loan_source_13',\n",
       " 'loan_source_999',\n",
       " 'loan_amount_usd',\n",
       " 'loan_use_1',\n",
       " 'loan_use_2',\n",
       " 'loan_use_3',\n",
       " 'loan_use_4',\n",
       " 'loan_use_5',\n",
       " 'loan_use_6',\n",
       " 'loan_use_7',\n",
       " 'loan_use_8',\n",
       " 'loan_use_9',\n",
       " 'loan_use_10',\n",
       " 'loan_use_11',\n",
       " 'loan_use_12',\n",
       " 'loan_use_13',\n",
       " 'loan_use_14',\n",
       " 'loan_use_15',\n",
       " 'loan_use_16',\n",
       " 'loan_use_17',\n",
       " 'loan_use_19',\n",
       " 'loan_use_20',\n",
       " 'loan_use_18',\n",
       " 'loan_use_999',\n",
       " 'distance_hometofarm',\n",
       " 'distance_localmarket',\n",
       " 'distance_inputsupplier',\n",
       " 'distance_healthclinic',\n",
       " 'distance_sealed_road',\n",
       " 'hh_own_house',\n",
       " 'hh_own_land',\n",
       " 'hh_sleep_room',\n",
       " 'hh_members_per_sleep_room',\n",
       " 'hh_drinking_water',\n",
       " 'hh_toilet',\n",
       " 'hh_cooking_fuel',\n",
       " 'hh_roof',\n",
       " 'hh_floor',\n",
       " 'hh_walls',\n",
       " 'hh_bed',\n",
       " 'hh_table',\n",
       " 'hh_cabinet_cupboard',\n",
       " 'hh_sofa',\n",
       " 'hh_watch',\n",
       " 'hh_wall_clock',\n",
       " 'hh_sewing_machine',\n",
       " 'hh_pots_cookware',\n",
       " 'hh_bicycle',\n",
       " 'hh_motorcycle_scooter',\n",
       " 'hh_animal_cart',\n",
       " 'hh_car_truck',\n",
       " 'hh_electricity',\n",
       " 'hh_radio',\n",
       " 'hh_bw_television',\n",
       " 'hh_color_television',\n",
       " 'hh_television',\n",
       " 'hh_mobile_telephone',\n",
       " 'hh_mobile_telephone_num',\n",
       " 'hh_telephone',\n",
       " 'hh_refrigerator',\n",
       " 'hh_freezer',\n",
       " 'hh_generator',\n",
       " 'hh_washing_machine',\n",
       " 'hh_iron',\n",
       " 'hh_computer_tablet',\n",
       " 'hh_camera',\n",
       " 'hh_video_dvd',\n",
       " 'hh_internet_access',\n",
       " 'hh_boat_motor',\n",
       " 'hh_boat_no_motor',\n",
       " 'hh_bank_account',\n",
       " 'hh_mosquito_net',\n",
       " 'hh_mosquito_net_num',\n",
       " 'hh_mosquito_net_lastnight',\n",
       " 'ls_dairy_cattle',\n",
       " 'ls_cows_bulls',\n",
       " 'ls_horses_donkeys_mules',\n",
       " 'ls_goats',\n",
       " 'ls_pigs',\n",
       " 'ls_rabbits',\n",
       " 'ls_grasscutter',\n",
       " 'ls_sheep',\n",
       " 'ls_chickens',\n",
       " 'ls_other_poultry',\n",
       " 'ls_other',\n",
       " 'hh_members_months_away',\n",
       " 'foodsec1',\n",
       " 'foodsec2',\n",
       " 'foodsec3',\n",
       " 'foodsec4',\n",
       " 'foodsec5',\n",
       " 'foodsec6',\n",
       " 'foodsec7',\n",
       " 'foodsec8',\n",
       " 'foodsec9',\n",
       " 'foodsec10',\n",
       " 'foodsec11',\n",
       " 'foodsec12',\n",
       " 'cocoa_hh_mostimp',\n",
       " 'migrant',\n",
       " 'youth',\n",
       " 'hh_income_1',\n",
       " 'hh_income_2',\n",
       " 'hh_income_3',\n",
       " 'hh_income_4',\n",
       " 'hh_income_5',\n",
       " 'hh_income_6',\n",
       " 'hh_income_7',\n",
       " 'hh_income_8',\n",
       " 'hh_income_9',\n",
       " 'hh_income_10',\n",
       " 'hh_income_11',\n",
       " 'hh_income_12',\n",
       " 'hh_income_13',\n",
       " 'n_hh_incomes',\n",
       " 'non_cocoa_land',\n",
       " 'share_cocoa_land',\n",
       " 'foodsec_bin_1',\n",
       " 'foodsec_bin_2',\n",
       " 'foodsec_bin_3',\n",
       " 'foodsec_bin_4',\n",
       " 'foodsec_bin_5',\n",
       " 'foodsec_bin_6',\n",
       " 'foodsec_bin_7',\n",
       " 'foodsec_bin_8',\n",
       " 'foodsec_bin_9',\n",
       " 'foodsec_bin_10',\n",
       " 'foodsec_bin_11',\n",
       " 'foodsec_bin_12',\n",
       " 'Analytical Group',\n",
       " 'Living Income Achieved']"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(data.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First some manual cases:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.rename(columns = {\"head_gender\" : \"Female head\"}, inplace = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.rename(columns = {\n",
    "     \"cocoa_training_source_1\": \"Training from farmer organisation\",\n",
    "    \"cocoa_training_source_2\": \"Training from government extension\",\n",
    "    \"cocoa_training_source_3\": \"Training from trader\",\n",
    "    \"cocoa_training_source_4\": \"Training from company, lbc\",\n",
    "    \"cocoa_training_source_5\": \"Training from ngo\",\n",
    "    \"cocoa_training_source_6\": \"Training from other\"\n",
    "}, inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now remaing cases using the stata dictionary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.rename(columns =  data_properties.variable_labels(), inplace = True )"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Country',\n",
       " 'Region',\n",
       " 'Number of household members living in the compound',\n",
       " 'Number of males over 65 years old',\n",
       " 'Number of females over 65 years old',\n",
       " 'Number of males 15-65',\n",
       " 'Number of females between 15 and 65 years old',\n",
       " 'Number of boys between 15 and 65 years old',\n",
       " 'Number of girls  between 5 and 14 years old',\n",
       " 'Number of boys younger than 4 years old',\n",
       " 'Number of girls younger than 4 years old',\n",
       " 'Dependency ratio',\n",
       " 'Head: age',\n",
       " 'Head: age group',\n",
       " 'Female head',\n",
       " 'Head: education level',\n",
       " 'All household boys ages 5 to 14 currently attending school',\n",
       " 'All household girls ages 5 to 14 currently attending school',\n",
       " \"Household's biggest source of income in last year\",\n",
       " \"Household's second biggest source of income in last year\",\n",
       " 'Estimated income percentage from sale of cocoa',\n",
       " 'Estimated income percentage from sale of other crops',\n",
       " 'Estimated income percentage from sale of livestock or livestock products',\n",
       " 'Estimated income percentage from sale of fish',\n",
       " 'Estimated income percentage from sale of bush products (bush meat, charcoal, woo',\n",
       " 'Estimated income percentage from labouring for other people on their farms',\n",
       " 'Estimated income percentage from ;abouring for other people non-agriculture',\n",
       " 'Estimated income percentage from own small business or trading',\n",
       " 'Estimated income percentage from salary employment with a company',\n",
       " 'Estimated income percentage from salary employment in government job (teacher, n',\n",
       " 'Estimated income percentage from sale or lease of land',\n",
       " 'Estimated income percentage from remittances from friends and family living away',\n",
       " 'Estimated income percentage from other',\n",
       " 'Produced avocado last season',\n",
       " 'Produced bananas last season',\n",
       " 'Produced soya beans last season',\n",
       " 'Produced beans last season',\n",
       " 'Produced cabbages last season',\n",
       " 'Produced carrots last season',\n",
       " 'Produced cashews last season',\n",
       " 'Produced cassava last season',\n",
       " 'Produced chili last season',\n",
       " 'Produced cocoa last season',\n",
       " 'Produced coconut last season',\n",
       " 'Produced cocoyam last season',\n",
       " 'Produced coffee last season',\n",
       " 'Produced courgette last season',\n",
       " 'Produced cotton last season',\n",
       " 'Produced cow pea last season',\n",
       " 'Produced eggplant last season',\n",
       " 'Produced fodder last season',\n",
       " 'Produced garlic last season',\n",
       " 'Produced ginger last season',\n",
       " 'Produced groundnuts last season',\n",
       " 'Produced lemon last season',\n",
       " 'Produced lime last season',\n",
       " 'Produced maize last season',\n",
       " 'Produced mango last season',\n",
       " 'Produced millet last season',\n",
       " 'Produced okra last season',\n",
       " 'Produced onions last season',\n",
       " 'Produced oranges last season',\n",
       " 'Produced palm last season',\n",
       " 'Produced papaya last season',\n",
       " 'Produced peas last season',\n",
       " 'Produced peppers last season',\n",
       " 'Produced pineapple last season',\n",
       " 'Produced plantain last season',\n",
       " 'Produced irish potato last season',\n",
       " 'Produced sweet potato last season',\n",
       " 'Produced rice last season',\n",
       " 'Produced rubber last season',\n",
       " 'Produced spinach last season',\n",
       " 'Produced sorghum last season',\n",
       " 'Produced shea nut last season',\n",
       " 'Produced sugar cane last season',\n",
       " 'Produced tangerine last season',\n",
       " 'Produced tomatoes last season',\n",
       " 'Produced watermelon last season',\n",
       " 'Produced yam last season',\n",
       " 'Produced other last season',\n",
       " 'Produced none/ not applicable last season',\n",
       " 'Number of crops produced',\n",
       " 'Sold avocado last season',\n",
       " 'Sold bananas last season',\n",
       " 'Sold soya beans last season',\n",
       " 'Sold beans last season',\n",
       " 'Sold cabbages last season',\n",
       " 'Sold carrots last season',\n",
       " 'Sold cashews last season',\n",
       " 'Sold cassava last season',\n",
       " 'Sold chili last season',\n",
       " 'Sold cocoa last season',\n",
       " 'Sold coconut last season',\n",
       " 'Sold cocoyam last season',\n",
       " 'Sold coffee last season',\n",
       " 'Sold courgette last season',\n",
       " 'Sold cotton last season',\n",
       " 'Sold cow pea last season',\n",
       " 'Sold eggplant last season',\n",
       " 'Sold fodder last season',\n",
       " 'Sold garlic last season',\n",
       " 'Sold ginger last season',\n",
       " 'Sold groundnuts last season',\n",
       " 'Sold lemon last season',\n",
       " 'Sold lime last season',\n",
       " 'Sold maize last season',\n",
       " 'Sold mango last season',\n",
       " 'Sold millet last season',\n",
       " 'Sold okra last season',\n",
       " 'Sold onions last season',\n",
       " 'Sold oranges last season',\n",
       " 'Sold palm last season',\n",
       " 'Sold papaya last season',\n",
       " 'Sold peas last season',\n",
       " 'Sold peppers last season',\n",
       " 'Sold pineapple last season',\n",
       " 'Sold plantain last season',\n",
       " 'Sold irish potato last season',\n",
       " 'Sold sweet potato last season',\n",
       " 'Sold rice last season',\n",
       " 'Sold rubber last season',\n",
       " 'Sold spinach last season',\n",
       " 'Sold sorghum last season',\n",
       " 'Sold shea nut last season',\n",
       " 'Sold sugar cane last season',\n",
       " 'Sold tangerine last season',\n",
       " 'Sold tomatoes last season',\n",
       " 'Sold watermelon last season',\n",
       " 'Sold yam last season',\n",
       " 'Sold other last season',\n",
       " 'Sold none/ not applicable last season',\n",
       " 'Sold number of crops sold last season',\n",
       " 'Second most important crop',\n",
       " 'Land used to cultivate all crops (ha)',\n",
       " 'All land owned (ha)',\n",
       " 'All land leased (ha)',\n",
       " 'All land under abunu sharecropping arrangement (ha)',\n",
       " 'All land under abusa sharecropping arrangement (ha)',\n",
       " 'All fallow land or land left for grazing (ha)',\n",
       " 'Total number of plots',\n",
       " 'Land leased out',\n",
       " 'Land leased out (ha)',\n",
       " 'Land under cocoa (ha)',\n",
       " 'Productive land under cocoa (ha)',\n",
       " 'Cocoa land owned (ha)',\n",
       " 'Cocoa land leased (ha) ',\n",
       " 'Cocoa land under abunu sharecropping arrangement (ha)',\n",
       " 'Cocoa land under abusa sharecropping arrangement (ha)',\n",
       " 'Number of cocoa plots',\n",
       " 'Years growing cocoa',\n",
       " 'Hectares with trees 5 years or younger',\n",
       " 'Hectares with trees between 5 and 25 years old',\n",
       " 'Hectares with trees older than 25 years old',\n",
       " 'Compared with 5 years ago, household increased, decreased, kept the same land',\n",
       " 'Cocoa is intercropped',\n",
       " 'Cocoa was intercropped with avocado',\n",
       " 'Cocoa was intercropped with bananas',\n",
       " 'Cocoa was intercropped with soya beans',\n",
       " 'Cocoa was intercropped with beans',\n",
       " 'Cocoa was intercropped with cabbages',\n",
       " 'Cocoa was intercropped with carrots',\n",
       " 'Cocoa was intercropped with cashews',\n",
       " 'Cocoa was intercropped with cassava',\n",
       " 'Cocoa was intercropped with chili',\n",
       " 'Cocoa was intercropped with cocoa',\n",
       " 'Cocoa was intercropped with coconut',\n",
       " 'Cocoa was intercropped with cocoyam',\n",
       " 'Cocoa was intercropped with coffee',\n",
       " 'Cocoa was intercropped with courgette',\n",
       " 'Cocoa was intercropped with cotton',\n",
       " 'Cocoa was intercropped with cow pea',\n",
       " 'Cocoa was intercropped with eggplant',\n",
       " 'Cocoa was intercropped with fodder',\n",
       " 'Cocoa was intercropped with garlic',\n",
       " 'Cocoa was intercropped with ginger',\n",
       " 'Cocoa was intercropped with groundnuts',\n",
       " 'Cocoa was intercropped with lemon',\n",
       " 'Cocoa was intercropped with lime',\n",
       " 'Cocoa was intercropped with maize',\n",
       " 'Cocoa was intercropped with mango',\n",
       " 'Cocoa was intercropped with millet',\n",
       " 'Cocoa was intercropped with okra',\n",
       " 'Cocoa was intercropped with onions',\n",
       " 'Cocoa was intercropped with oranges',\n",
       " 'Cocoa was intercropped with palm',\n",
       " 'Cocoa was intercropped with papaya',\n",
       " 'Cocoa was intercropped with peas',\n",
       " 'Cocoa was intercropped with peppers',\n",
       " 'Cocoa was intercropped with pineapple',\n",
       " 'Cocoa was intercropped with plantain',\n",
       " 'Cocoa was intercropped with irish potato',\n",
       " 'Cocoa was intercropped with sweet potato',\n",
       " 'Cocoa was intercropped with rice',\n",
       " 'Cocoa was intercropped with rubber',\n",
       " 'Cocoa was intercropped with spinach',\n",
       " 'Cocoa was intercropped with sorghum',\n",
       " 'Cocoa was intercropped with shea nut',\n",
       " 'Cocoa was intercropped with sugar cane',\n",
       " 'Cocoa was intercropped with tangerine',\n",
       " 'Cocoa was intercropped with tomatoes',\n",
       " 'Cocoa was intercropped with watermelon',\n",
       " 'Cocoa was intercropped with yam',\n",
       " 'Cocoa was intercropped with other',\n",
       " 'Cocoa was intercropped with none/ not applicable',\n",
       " 'Cocoa production (kg)',\n",
       " 'Cocoa production (kg/ha)',\n",
       " 'Cocoa variety: tetteh quarshie',\n",
       " 'Cocoa variety: amazonia',\n",
       " 'Cocoa variety: hybrid',\n",
       " 'Cocoa variety: hybride (2nd generation)',\n",
       " 'Cocoa variety: other',\n",
       " 'Cocoa variety: french',\n",
       " 'Cocoa variety: tout venant',\n",
       " \"Cocoa variety: don't know\",\n",
       " 'Trees are planted in rows',\n",
       " 'Land clearing for cocoa was done last year',\n",
       " 'Land clearing done by household labour',\n",
       " 'Land clearing done by hired labour',\n",
       " 'Land clearing done by communal labour',\n",
       " 'Land clearing done by company/codapec labour',\n",
       " 'Land clearing was done by man',\n",
       " 'Land clearing was done by women',\n",
       " 'Land preparation for cocoa was done last year',\n",
       " 'Land preparation done by household labour',\n",
       " 'Land preparation done by hired labour',\n",
       " 'Land preparation done by communal labour',\n",
       " 'Land preparation done by company/codapec labour',\n",
       " 'Land preparation was done by man',\n",
       " 'Land preparation was done by women',\n",
       " 'Cocoa planting was done last year',\n",
       " 'Planting done by household labour',\n",
       " 'Planting done by hired labour',\n",
       " 'Planting done by communal labour',\n",
       " 'Planting done by company/codapec labour',\n",
       " 'Planting was done by man',\n",
       " 'Planting was done by women',\n",
       " 'Seeds source: own farm',\n",
       " 'Seeds source: neighbours / friends / relatives farms',\n",
       " 'Seeds source: farmer organisation',\n",
       " 'Seeds source: government, extension',\n",
       " 'Seeds source: purchased from agrodealer shop',\n",
       " 'Seeds source: trader',\n",
       " 'Seeds source: company / lbc',\n",
       " 'Seeds source: other',\n",
       " 'Seeds source: nestle, cargill, olam, ecom other company',\n",
       " 'Seeds source: ong',\n",
       " \"Seeds source: don't know\",\n",
       " 'Granular fertilizer was applied on cocoa land',\n",
       " 'Granular fertilizer application done by household labour',\n",
       " 'Granular fertilizer application done by hired labour',\n",
       " 'Granular fertilizer application done by communal labour',\n",
       " 'Granular fertilizer application done by company/codapec labour',\n",
       " 'Granular fertilizer application was done by man',\n",
       " 'Granular fertilizer application was done by women',\n",
       " 'Liquid fertilizer was applied on cocoa land',\n",
       " 'Liquid fertilizer application done by household labour',\n",
       " 'Liquid fertilizer application done by hired labour',\n",
       " 'Liquid fertilizer application done by communal labour',\n",
       " 'Liquid fertilizer application done by company/codapec labour',\n",
       " 'Liquid fertilizer application was done by man',\n",
       " 'Liquid fertilizer application was done by women',\n",
       " 'Manure or compost was applied on cocoa land',\n",
       " 'Manure or compost application done by household labour',\n",
       " 'Manure or compost application done by hired labour',\n",
       " 'Manure or compost application done by communal labour',\n",
       " 'Manure or compost application done by company/codapec labour',\n",
       " 'Manure or compost application was done by man',\n",
       " 'Manure or compost application was done by women',\n",
       " 'Herbicides was applied on cocoa land',\n",
       " 'Herbicide application done by household labour',\n",
       " 'Herbicide application done by hired labour',\n",
       " 'Herbicide application done by communal labour',\n",
       " 'Herbicide application done by company/codapec labour',\n",
       " 'Herbicide application was done by man',\n",
       " 'Herbicide application was done by women',\n",
       " 'Pesticides was applied on cocoa land',\n",
       " 'Pesticide application done by household labour',\n",
       " 'Pesticide application done by hired labour',\n",
       " 'Pesticide application done by communal labour',\n",
       " 'Pesticide application done by company/codapec labour',\n",
       " 'Pesticide application was done by man',\n",
       " 'Pesticide application was done by women',\n",
       " 'Fungicides was applied on cocoa land',\n",
       " 'Fungicide application done by household labour',\n",
       " 'Fungicide application done by hired labour',\n",
       " 'Fungicide application done by communal labour',\n",
       " 'Fungicide application done by company/codapec labour',\n",
       " 'Fungicide application was done by man',\n",
       " 'Fungicide application was done by women',\n",
       " 'Weeding was done on cocoa land',\n",
       " 'Weeding done by household labour',\n",
       " 'Weeding done by hired labour',\n",
       " 'Weeding done by communal labour',\n",
       " 'Weeding done by company/codapec labour',\n",
       " 'Weeding was done by man',\n",
       " 'Weeding was done by women',\n",
       " 'Pruuning was done on cocoa land',\n",
       " 'Prunning done by household labour',\n",
       " 'Prunning done by hired labour',\n",
       " 'Prunning done by communal labour',\n",
       " 'Prunning done by company/codapec labour',\n",
       " 'Prunning was done by man',\n",
       " 'Prunning was done by women',\n",
       " 'Cocoa was harvested last year',\n",
       " 'Harvesting: january',\n",
       " 'Harvesting: february',\n",
       " 'Harvesting: march',\n",
       " 'Harvesting: april',\n",
       " 'Harvesting: may',\n",
       " 'Harvesting: june',\n",
       " 'Harvesting: july',\n",
       " 'Harvesting: august',\n",
       " 'Harvesting: september',\n",
       " 'Harvesting: october',\n",
       " 'Harvesting: november',\n",
       " 'Harvesting: december',\n",
       " \"Harvesting: don't know\",\n",
       " 'Harvesting: none/ not applicable',\n",
       " 'Harvesting done by household labour',\n",
       " 'Harvesting done by hired labour',\n",
       " 'Harvesting done by communal labour',\n",
       " 'Harvesting done by company/codapec labour',\n",
       " 'Harvesting was done by man',\n",
       " 'Harvesting was done by women',\n",
       " 'Cocoa pod breaking was done last year',\n",
       " 'Pod breaking done by household labour',\n",
       " 'Pod breaking done by hired labour',\n",
       " 'Pod breaking done by communal labour',\n",
       " 'Pod breaking done by company/codapec labour',\n",
       " 'Pod breaking was done by man',\n",
       " 'Pod breaking was done by women',\n",
       " 'Cocoa fermenting was done last year',\n",
       " 'Fermenting done by household labour',\n",
       " 'Fermenting done by hired labour',\n",
       " 'Fermenting done by communal labour',\n",
       " 'Fermenting done by company/codapec labour',\n",
       " 'Fermenting was done by man',\n",
       " 'Fermenting was done by women',\n",
       " 'Cocoa transported from the farm to the homestead',\n",
       " 'Transporting done by household labour',\n",
       " 'Transporting done by hired labour',\n",
       " 'Transporting done by communal labour',\n",
       " 'Transporting done by company/codapec labour',\n",
       " 'Transporting was done by man',\n",
       " 'Transporting was done by women',\n",
       " 'Cocoa was dried last year',\n",
       " 'Drying done by household labour',\n",
       " 'Drying done by hired labour',\n",
       " 'Drying done by communal labour',\n",
       " 'Drying done by company/codapec labour',\n",
       " 'Drying was done by man',\n",
       " 'Drying was done by women',\n",
       " 'Cocoa was lost to dieseases',\n",
       " 'Cocoa was lost to pests',\n",
       " 'Cocoa was lost during storage',\n",
       " 'Cocoa disease type: cocoa swollen shoot virus disease (cssvd)',\n",
       " 'Cocoa disease type: black pod disease',\n",
       " 'Cocoa disease type: blight thread',\n",
       " 'Cocoa disease type: le chancre',\n",
       " 'Cocoa disease type: other',\n",
       " 'Cocoa pest type: mistletoe',\n",
       " 'Cocoa pest type: akate',\n",
       " 'Cocoa pest type: mirids',\n",
       " 'Cocoa pest type: black ants',\n",
       " 'Cocoa pest type: termites',\n",
       " 'Cocoa pest type: rodents',\n",
       " 'Cocoa pest type: other',\n",
       " 'Cocoa pest type: foreuse de tige',\n",
       " 'Cocoa pest type: mousse verte',\n",
       " \"Cocoa pest type: don't know\",\n",
       " 'Cocoa records were kept',\n",
       " 'Cocoa records: money spent on inputs',\n",
       " 'Cocoa records: money spent on hired labourers',\n",
       " 'Cocoa records: number of days household members spent labouring',\n",
       " 'Cocoa records: the volume produced',\n",
       " 'Cocoa records: the volume sold, the price and money received',\n",
       " \"Cocoa records: don't know\",\n",
       " 'Decision making was done by man',\n",
       " 'Decision making was done by women',\n",
       " 'Cocoa sold (share of total)',\n",
       " 'Cocoa consumed at home (share of total)',\n",
       " 'Cocoa duty abunu/abusan (share of total)',\n",
       " 'Cocoa given away free (share of total)',\n",
       " 'Cocoa given in exchange for other goods (barter) (share of total)',\n",
       " 'Cocoa given as payment for labour (share of total)',\n",
       " 'Cocoa lost (eg from bad quality, rotting, theft) (share of total)',\n",
       " 'Country average cocoa price (usd/kg)',\n",
       " 'Sales of cocoa in: january',\n",
       " 'Sales of cocoa in: february',\n",
       " 'Sales of cocoa in: march',\n",
       " 'Sales of cocoa in: april',\n",
       " 'Sales of cocoa in: may',\n",
       " 'Sales of cocoa in: june',\n",
       " 'Sales of cocoa in: july',\n",
       " 'Sales of cocoa in: august',\n",
       " 'Sales of cocoa in: september',\n",
       " 'Sales of cocoa in: october',\n",
       " 'Sales of cocoa in: november',\n",
       " 'Sales of cocoa in: december',\n",
       " \"Sales of cocoa in: don't know\",\n",
       " 'Sales of cocoa in: none/ not applicable',\n",
       " 'Cocoa sold to pbc/produce buying company',\n",
       " 'Cocoa sold to aamc/akuafo adamfo',\n",
       " 'Cocoa sold to abl/adwumapa',\n",
       " 'Cocoa sold to olam',\n",
       " 'Cocoa sold to fcl/fedco',\n",
       " 'Cocoa sold to agl/armajaro',\n",
       " 'Cocoa sold to tgl/transroyal',\n",
       " 'Cocoa sold to kkl/kuapa kokoo',\n",
       " 'Cocoa sold to cmgl/cocoa merchants',\n",
       " 'Cocoa sold to rcl/royal commodities',\n",
       " 'Cocoa sold to cemoi',\n",
       " 'Cocoa sold to barry callebault / saco',\n",
       " 'Cocoa sold to cargill',\n",
       " 'Cocoa sold to coop곡tive/groupement/association de producteurs',\n",
       " 'Cocoa sold to other licenced buying company (lbc)',\n",
       " 'Cocoa sold to other non-licenced buyer',\n",
       " 'Cocoa sold to other',\n",
       " \"Cocoa sold to don't know\",\n",
       " 'Cocoa was sold by man',\n",
       " 'Cocoa was sold by women',\n",
       " 'Received training in the last 5 years',\n",
       " 'Number of training days related to cocoa in the past 5 years',\n",
       " 'If training was received in the last 5 years, was training received in the last ',\n",
       " 'Training on good production practices',\n",
       " 'Training on good environmental practices',\n",
       " 'Training on pest and disease management',\n",
       " 'Training on post-harvest storage and processing',\n",
       " 'Training on health and safety',\n",
       " 'Training on gender',\n",
       " 'Training on child labour',\n",
       " 'Training on record keeping and business management',\n",
       " 'Training on savings and loans',\n",
       " 'Training on cooperative management',\n",
       " 'Training on other',\n",
       " 'Training from farmer organisation',\n",
       " 'Training from government extension',\n",
       " 'Training from trader',\n",
       " 'Training from company, lbc',\n",
       " 'Training from ngo',\n",
       " 'Training from other',\n",
       " \"Training on don't know\",\n",
       " 'Certified by utz, ra, fairtrade or organic (removing other values)',\n",
       " 'Certification: utz',\n",
       " 'Certification: rainforest alliance',\n",
       " 'Certification: flo / fairtrade',\n",
       " 'Certification: organic',\n",
       " 'Household is member of cocoa producer group',\n",
       " 'Producer group services: trainings',\n",
       " 'Producer group services: providing inputs (eg fertilizer, planting material)',\n",
       " 'Producer group services: savings facility',\n",
       " 'Producer group services: loans facility',\n",
       " 'Producer group services: organising labour',\n",
       " 'Producer group services: storage of produce',\n",
       " 'Producer group services: grading of produce',\n",
       " 'Producer group services: marketing',\n",
       " 'Producer group services: other',\n",
       " 'Producer group services: none',\n",
       " \"Producer group services: don't know\",\n",
       " 'Received inputs on credit',\n",
       " 'Inputs on credit: fertilizer',\n",
       " 'Inputs on credit: pesticides, herbicides or fungicides',\n",
       " 'Inputs on credit: seedlings',\n",
       " 'Inputs on credit: other',\n",
       " 'Money borrowed last year',\n",
       " 'Loan source: family or friends',\n",
       " 'Loan source: village money lender',\n",
       " 'Loan source: lbc or pc (licenced buying company or purchasing clerk)',\n",
       " 'Loan source: vsla/islc (informal savings and loans group)',\n",
       " 'Loan source: savings and credit cooperative (sacco) / credit union',\n",
       " 'Loan source: microfinance institution (mfi)',\n",
       " 'Loan source: bank',\n",
       " 'Loan source: trader',\n",
       " 'Loan source: company',\n",
       " 'Loan source: church / mosque',\n",
       " 'Loan source: other',\n",
       " 'Loan source: pisteurs',\n",
       " \"Loan source: cooperative, producer's association, group\",\n",
       " \"Loan source: don't know\",\n",
       " 'Loan Amounts (USD)',\n",
       " 'Loan use: seedlings',\n",
       " 'Loan use: inputs (fertilizer, herbicides, pesticides)',\n",
       " 'Loan use: hired labourers',\n",
       " 'Loan use: farm tools',\n",
       " 'Loan use: hire machinary, tractor',\n",
       " 'Loan use: machinary, tractor',\n",
       " 'Loan use: car or motorbike',\n",
       " 'Loan use: buy land or lease land',\n",
       " 'Loan use: transport costs',\n",
       " 'Loan use: storage',\n",
       " 'Loan use: repair machinary',\n",
       " 'Loan use: processing',\n",
       " 'Loan use: farmer organisation membership fees',\n",
       " 'Loan use: school fees',\n",
       " 'Loan use: clothes',\n",
       " 'Loan use: food for consumption',\n",
       " 'Loan use: funeral costs',\n",
       " 'Loan use: wedding costs',\n",
       " 'Loan use: small business / trade',\n",
       " 'Loan use: other',\n",
       " \"Loan use: don't know\",\n",
       " 'Distance to your main farm',\n",
       " 'Distance to the nearest local market',\n",
       " 'Distance to the nearest input supplier?',\n",
       " 'Distance to the nearest health clinic',\n",
       " 'Distance to the nearest tarred road (sealed road)',\n",
       " 'House owned',\n",
       " 'Land owned',\n",
       " 'Number of sleeping rooms',\n",
       " 'Members per sleeping rooms',\n",
       " 'Main source of drinking water',\n",
       " 'Type of toilet',\n",
       " 'Cooking fuel',\n",
       " 'Roof material',\n",
       " 'Floor material',\n",
       " 'Walls material',\n",
       " 'Bed',\n",
       " 'Table',\n",
       " 'Cabinet/cupboard',\n",
       " 'sofa / couch',\n",
       " 'Watch',\n",
       " 'Wall clock',\n",
       " 'Sewing machine',\n",
       " 'Pots and cookware',\n",
       " 'Bicycle',\n",
       " 'Motorcycle or Scooter',\n",
       " 'Animal-drawn cart',\n",
       " 'Car or Truck',\n",
       " 'Electricity',\n",
       " 'Radio',\n",
       " 'Black & white television',\n",
       " 'Color television',\n",
       " 'Television (BW or Color)',\n",
       " 'Mobile telephone',\n",
       " 'Number of mobile phones in the household',\n",
       " 'Telephone (non-mobile)',\n",
       " 'Refrigerator',\n",
       " 'Freezer',\n",
       " 'Generator',\n",
       " 'Washing machine',\n",
       " 'Box or electric iron',\n",
       " 'Computer/tablet computer',\n",
       " 'Photo camera ( not on phone )',\n",
       " 'Video deck/DVD/VCD',\n",
       " 'Internet access',\n",
       " 'Boat with a motor',\n",
       " 'Boat without a motor',\n",
       " 'Bank account',\n",
       " 'Mosquito net',\n",
       " 'Number of mosquito nets in your household',\n",
       " 'Did all members of your household sleep under a mosquito net last night?',\n",
       " 'Dairy cattle',\n",
       " 'Cows/bulls',\n",
       " 'Horses/donkeys/mules',\n",
       " 'Goats',\n",
       " 'Pigs',\n",
       " 'Rabbits',\n",
       " 'Grasscutter',\n",
       " 'Sheep',\n",
       " 'Chickens',\n",
       " 'Other poultry',\n",
       " 'Other animals',\n",
       " 'Number of members how in living away',\n",
       " 'Meals per day: january',\n",
       " 'Meals per day: february',\n",
       " 'Meals per day: march',\n",
       " 'Meals per day: april',\n",
       " 'Meals per day: may',\n",
       " 'Meals per day: june',\n",
       " 'Meals per day: july',\n",
       " 'Meals per day: august',\n",
       " 'Meals per day: september',\n",
       " 'Meals per day: october',\n",
       " 'Meals per day: november',\n",
       " 'Meals per day: december',\n",
       " 'Cocoa Farmer (Importance)',\n",
       " 'Migrant',\n",
       " 'Youth',\n",
       " 'Household income source: sale of cocoa',\n",
       " 'Household income source: sale of other crops',\n",
       " 'Household income source: sale of livestock or livestock products',\n",
       " 'Household income source: sale of fish',\n",
       " 'Household income source: sale of bush products (bush meat, charcoal, wood etc)',\n",
       " 'Household income source: labouring for other people on their farms',\n",
       " 'Household income source: labouring for other people non-agriculture',\n",
       " 'Household income source: own small business or trading',\n",
       " 'Household income source: salary employment with a company',\n",
       " 'Household income source: salary employment in government job (teacher, nurse, po',\n",
       " 'Household income source: sale or lease of land',\n",
       " 'Household income source: remittances - money given by friends and family who are',\n",
       " 'Household income source: other',\n",
       " 'Number of different household sources of incomes',\n",
       " 'Land (ha) used for non-cocoa',\n",
       " 'Share of cocoa land over all land',\n",
       " '3 meals/day: january',\n",
       " '3 meals/day: february',\n",
       " '3 meals/day: march',\n",
       " '3 meals/day: april',\n",
       " '3 meals/day: may',\n",
       " '3 meals/day: june',\n",
       " '3 meals/day: july',\n",
       " '3 meals/day: august',\n",
       " '3 meals/day: september',\n",
       " '3 meals/day: october',\n",
       " '3 meals/day: november',\n",
       " '3 meals/day: december',\n",
       " 'Analytical Group',\n",
       " 'Living Income Achieved']"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list(data.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's do a final check per country:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ghana    970\n",
       "CdI      407\n",
       "Name: Country, dtype: int64"
      ]
     },
     "execution_count": 122,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Country'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Final check overall:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Number of household members living in the compound</th>\n",
       "      <th>Number of males over 65 years old</th>\n",
       "      <th>Number of females over 65 years old</th>\n",
       "      <th>Number of males 15-65</th>\n",
       "      <th>Number of females between 15 and 65 years old</th>\n",
       "      <th>Number of boys between 15 and 65 years old</th>\n",
       "      <th>Number of girls  between 5 and 14 years old</th>\n",
       "      <th>Number of boys younger than 4 years old</th>\n",
       "      <th>Number of girls younger than 4 years old</th>\n",
       "      <th>Dependency ratio</th>\n",
       "      <th>...</th>\n",
       "      <th>3 meals/day: march</th>\n",
       "      <th>3 meals/day: april</th>\n",
       "      <th>3 meals/day: may</th>\n",
       "      <th>3 meals/day: june</th>\n",
       "      <th>3 meals/day: july</th>\n",
       "      <th>3 meals/day: august</th>\n",
       "      <th>3 meals/day: september</th>\n",
       "      <th>3 meals/day: october</th>\n",
       "      <th>3 meals/day: november</th>\n",
       "      <th>3 meals/day: december</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1365.000000</td>\n",
       "      <td>1365.000000</td>\n",
       "      <td>1365.000000</td>\n",
       "      <td>1365.000000</td>\n",
       "      <td>1365.000000</td>\n",
       "      <td>1365.000000</td>\n",
       "      <td>1365.000000</td>\n",
       "      <td>1365.000000</td>\n",
       "      <td>1365.000000</td>\n",
       "      <td>1349.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1377.000000</td>\n",
       "      <td>1377.000000</td>\n",
       "      <td>1377.000000</td>\n",
       "      <td>1377.000000</td>\n",
       "      <td>1377.000000</td>\n",
       "      <td>1377.000000</td>\n",
       "      <td>1377.000000</td>\n",
       "      <td>1377.000000</td>\n",
       "      <td>1377.000000</td>\n",
       "      <td>1377.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>6.190476</td>\n",
       "      <td>0.133333</td>\n",
       "      <td>0.109158</td>\n",
       "      <td>1.734799</td>\n",
       "      <td>1.688645</td>\n",
       "      <td>0.979487</td>\n",
       "      <td>0.848352</td>\n",
       "      <td>0.354579</td>\n",
       "      <td>0.342125</td>\n",
       "      <td>1.011092</td>\n",
       "      <td>...</td>\n",
       "      <td>0.683370</td>\n",
       "      <td>0.674655</td>\n",
       "      <td>0.672476</td>\n",
       "      <td>0.662309</td>\n",
       "      <td>0.657952</td>\n",
       "      <td>0.661583</td>\n",
       "      <td>0.697168</td>\n",
       "      <td>0.708061</td>\n",
       "      <td>0.712418</td>\n",
       "      <td>0.724038</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>3.002501</td>\n",
       "      <td>0.369010</td>\n",
       "      <td>0.351719</td>\n",
       "      <td>1.221747</td>\n",
       "      <td>1.148336</td>\n",
       "      <td>1.096323</td>\n",
       "      <td>1.055501</td>\n",
       "      <td>0.671139</td>\n",
       "      <td>0.656139</td>\n",
       "      <td>1.011952</td>\n",
       "      <td>...</td>\n",
       "      <td>0.465331</td>\n",
       "      <td>0.468674</td>\n",
       "      <td>0.469481</td>\n",
       "      <td>0.473094</td>\n",
       "      <td>0.474568</td>\n",
       "      <td>0.473343</td>\n",
       "      <td>0.459650</td>\n",
       "      <td>0.454820</td>\n",
       "      <td>0.452799</td>\n",
       "      <td>0.447160</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>4.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>6.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.500000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>21.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 566 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       Number of household members living in the compound  \\\n",
       "count                                        1365.000000    \n",
       "mean                                            6.190476    \n",
       "std                                             3.002501    \n",
       "min                                             1.000000    \n",
       "25%                                             4.000000    \n",
       "50%                                             6.000000    \n",
       "75%                                             8.000000    \n",
       "max                                            21.000000    \n",
       "\n",
       "       Number of males over 65 years old  Number of females over 65 years old  \\\n",
       "count                        1365.000000                          1365.000000   \n",
       "mean                            0.133333                             0.109158   \n",
       "std                             0.369010                             0.351719   \n",
       "min                             0.000000                             0.000000   \n",
       "25%                             0.000000                             0.000000   \n",
       "50%                             0.000000                             0.000000   \n",
       "75%                             0.000000                             0.000000   \n",
       "max                             5.000000                             5.000000   \n",
       "\n",
       "       Number of males 15-65  Number of females between 15 and 65 years old  \\\n",
       "count            1365.000000                                    1365.000000   \n",
       "mean                1.734799                                       1.688645   \n",
       "std                 1.221747                                       1.148336   \n",
       "min                 0.000000                                       0.000000   \n",
       "25%                 1.000000                                       1.000000   \n",
       "50%                 1.000000                                       1.000000   \n",
       "75%                 2.000000                                       2.000000   \n",
       "max                 8.000000                                       9.000000   \n",
       "\n",
       "       Number of boys between 15 and 65 years old  \\\n",
       "count                                 1365.000000   \n",
       "mean                                     0.979487   \n",
       "std                                      1.096323   \n",
       "min                                      0.000000   \n",
       "25%                                      0.000000   \n",
       "50%                                      1.000000   \n",
       "75%                                      2.000000   \n",
       "max                                      6.000000   \n",
       "\n",
       "       Number of girls  between 5 and 14 years old  \\\n",
       "count                                  1365.000000   \n",
       "mean                                      0.848352   \n",
       "std                                       1.055501   \n",
       "min                                       0.000000   \n",
       "25%                                       0.000000   \n",
       "50%                                       1.000000   \n",
       "75%                                       1.000000   \n",
       "max                                       6.000000   \n",
       "\n",
       "       Number of boys younger than 4 years old  \\\n",
       "count                              1365.000000   \n",
       "mean                                  0.354579   \n",
       "std                                   0.671139   \n",
       "min                                   0.000000   \n",
       "25%                                   0.000000   \n",
       "50%                                   0.000000   \n",
       "75%                                   1.000000   \n",
       "max                                   5.000000   \n",
       "\n",
       "       Number of girls younger than 4 years old  Dependency ratio  ...  \\\n",
       "count                               1365.000000       1349.000000  ...   \n",
       "mean                                   0.342125          1.011092  ...   \n",
       "std                                    0.656139          1.011952  ...   \n",
       "min                                    0.000000          0.000000  ...   \n",
       "25%                                    0.000000          0.333333  ...   \n",
       "50%                                    0.000000          0.750000  ...   \n",
       "75%                                    1.000000          1.500000  ...   \n",
       "max                                    5.000000         13.000000  ...   \n",
       "\n",
       "       3 meals/day: march  3 meals/day: april  3 meals/day: may  \\\n",
       "count         1377.000000         1377.000000       1377.000000   \n",
       "mean             0.683370            0.674655          0.672476   \n",
       "std              0.465331            0.468674          0.469481   \n",
       "min              0.000000            0.000000          0.000000   \n",
       "25%              0.000000            0.000000          0.000000   \n",
       "50%              1.000000            1.000000          1.000000   \n",
       "75%              1.000000            1.000000          1.000000   \n",
       "max              1.000000            1.000000          1.000000   \n",
       "\n",
       "       3 meals/day: june  3 meals/day: july  3 meals/day: august  \\\n",
       "count        1377.000000        1377.000000          1377.000000   \n",
       "mean            0.662309           0.657952             0.661583   \n",
       "std             0.473094           0.474568             0.473343   \n",
       "min             0.000000           0.000000             0.000000   \n",
       "25%             0.000000           0.000000             0.000000   \n",
       "50%             1.000000           1.000000             1.000000   \n",
       "75%             1.000000           1.000000             1.000000   \n",
       "max             1.000000           1.000000             1.000000   \n",
       "\n",
       "       3 meals/day: september  3 meals/day: october  3 meals/day: november  \\\n",
       "count             1377.000000           1377.000000            1377.000000   \n",
       "mean                 0.697168              0.708061               0.712418   \n",
       "std                  0.459650              0.454820               0.452799   \n",
       "min                  0.000000              0.000000               0.000000   \n",
       "25%                  0.000000              0.000000               0.000000   \n",
       "50%                  1.000000              1.000000               1.000000   \n",
       "75%                  1.000000              1.000000               1.000000   \n",
       "max                  1.000000              1.000000               1.000000   \n",
       "\n",
       "       3 meals/day: december  \n",
       "count            1377.000000  \n",
       "mean                0.724038  \n",
       "std                 0.447160  \n",
       "min                 0.000000  \n",
       "25%                 0.000000  \n",
       "50%                 1.000000  \n",
       "75%                 1.000000  \n",
       "max                 1.000000  \n",
       "\n",
       "[8 rows x 566 columns]"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Export"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [],
   "source": [
    "data.to_pickle('data_ready_for_ML.pkl')"
   ]
  },
  {
   "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": "3.7.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}