{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "5cfdceec",
   "metadata": {},
   "source": [
    "# Independent samples t-test"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "741220b6",
   "metadata": {},
   "source": [
    "### Set up Python libraries\n",
    "\n",
    "As usual, run the code cell below to import the relevant Python libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "692abf91",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Set-up Python libraries - you need to run this but you don't need to change it\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import scipy.stats as stats\n",
    "import pandas \n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "38124844",
   "metadata": {},
   "source": [
    "## Example\n",
    "\n",
    "<img src=\"https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook/main/images/appletree.jpg\" width=30% >\n",
    "\n",
    "\n",
    "Below we have data on the average weight, in grams, of apples from each of 20 apple trees.\n",
    "\n",
    "10 of the trees received MiracleGro fertilizer and the other 10 received Brand X.\n",
    "\n",
    "Test the hypothesis that the trees given MiracleGro produced heavier apples."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f2ba6e37",
   "metadata": {},
   "source": [
    "### Inspect the data\n",
    "\n",
    "The data are provided in a text (.csv) file.\n",
    "\n",
    "Let's load the data as a Pandas dataframe, and plot them to get a sense for their distribution (is it normal?) and any outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "3f184f9d",
   "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>Fertilizer</th>\n",
       "      <th>meanAppleWeight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>176</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Fertilizer  meanAppleWeight\n",
       "0       BrandX              172\n",
       "1       BrandX              165\n",
       "2       BrandX              175\n",
       "3       BrandX              164\n",
       "4       BrandX              165\n",
       "5       BrandX              157\n",
       "6       BrandX              183\n",
       "7       BrandX              186\n",
       "8       BrandX              191\n",
       "9       BrandX              173\n",
       "10  MiracleGro              164\n",
       "11  MiracleGro              198\n",
       "12  MiracleGro              184\n",
       "13  MiracleGro              200\n",
       "14  MiracleGro              180\n",
       "15  MiracleGro              189\n",
       "16  MiracleGro              177\n",
       "17  MiracleGro              170\n",
       "18  MiracleGro              192\n",
       "19  MiracleGro              193\n",
       "20  MiracleGro              187\n",
       "21  MiracleGro              176"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# load the data and have a look\n",
    "pandas.read_csv('https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook/main/data/AppleWeights.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ce8aef6",
   "metadata": {},
   "source": [
    "Let's plot the data and see if they look Normally distributed.\n",
    "\n",
    "As we saw in the session on plotting, a good choice here will be a KDE plot (to get an estimate of the shape of the distribution) and a rug plot (individual data values as the KDE plot is based on only a small sample)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5b4ca4f7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "apples = pandas.read_csv('https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook/main/data/AppleWeights.csv')\n",
    "\n",
    "# let's make separate dataframes for the two brands of fertilizer\n",
    "apples_BrandX = apples[apples[\"Fertilizer\"]==\"BrandX\"]\n",
    "apples_MiracleGro = apples[apples[\"Fertilizer\"]==\"MiracleGro\"]\n",
    "\n",
    "sns.kdeplot(apples_BrandX[\"meanAppleWeight\"], color='b', shade=True, label='BrandX')\n",
    "sns.kdeplot(apples_MiracleGro[\"meanAppleWeight\"], color='r', shade=True, label='MiracleGro')\n",
    "\n",
    "sns.rugplot(apples_BrandX[\"meanAppleWeight\"], color='b')\n",
    "sns.rugplot(apples_MiracleGro[\"meanAppleWeight\"], color='r')\n",
    "\n",
    "plt.xlabel(\"Mean weight of apples (g)\", fontsize = 12)\n",
    "plt.ylabel(\"Density\", fontsize = 12)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "08178c3d",
   "metadata": {},
   "source": [
    "It looks like both distributions are unimodal but a bit skewed. \n",
    "\n",
    "Typically, it would be hard to say if the data are really non-normal due to the small number \n",
    "of data points (the apparent skew could be a fluke). By default, I would err on the side of caution and use a non-parametric test.\n",
    "\n",
    "However, in this particular case I am confident that the data in each sample are drawn from a normal distribution \n",
    "because each data point represents the mean of a large sample (the weights of all apples on one tree) \n",
    "and such means are always normally distributed due to the Central Limit theorem.\n",
    "There is a video explaining why here.\n",
    "\n",
    "So we can go ahead and use a t-test."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "97b80d53",
   "metadata": {},
   "source": [
    "### Hypotheses\n",
    "\n",
    "Ho: the mean weight of apples is the same for trees fertilized with MiracleGro and Brand X\n",
    "\n",
    "Ha: the mean weight of apples is greater for trees fertilized with MiracleGro\n",
    "    \n",
    "This is a one tailed test as the manufacturers of MiracleGro are only looking for an effect in one direction \n",
    "(evidence that MiracleGro is better!)\n",
    "\n",
    "We will test at the $\\alpha = 0.05$ significance level"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b79fc39f",
   "metadata": {},
   "source": [
    "### Descriptive statistics\n",
    "\n",
    "First, we obtain the relevant desriptive statistics. By relevant, I mean the ones that go into the equation for the t-test:\n",
    "\n",
    "\n",
    "$$ t = \\frac{\\bar{x_1} - \\bar{x_2}}{s \\sqrt{\\frac{1}{n_1}+\\frac{1}{n_2}}} $$\n",
    "\n",
    "This would be the sample means $\\bar{x_1}$ and $\\bar{x_2}$, and the standard deviations for both samples (these feed into the pooled standard deviation $s$) and the sample size $n$.\n",
    "\n",
    "Remember, <tt>apples</tt> is our original Pandas dataframe:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "4825c72f",
   "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>Fertilizer</th>\n",
       "      <th>meanAppleWeight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>165</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>191</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>BrandX</td>\n",
       "      <td>173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>164</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>198</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>200</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>180</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>189</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>193</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>187</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>MiracleGro</td>\n",
       "      <td>176</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    Fertilizer  meanAppleWeight\n",
       "0       BrandX              172\n",
       "1       BrandX              165\n",
       "2       BrandX              175\n",
       "3       BrandX              164\n",
       "4       BrandX              165\n",
       "5       BrandX              157\n",
       "6       BrandX              183\n",
       "7       BrandX              186\n",
       "8       BrandX              191\n",
       "9       BrandX              173\n",
       "10  MiracleGro              164\n",
       "11  MiracleGro              198\n",
       "12  MiracleGro              184\n",
       "13  MiracleGro              200\n",
       "14  MiracleGro              180\n",
       "15  MiracleGro              189\n",
       "16  MiracleGro              177\n",
       "17  MiracleGro              170\n",
       "18  MiracleGro              192\n",
       "19  MiracleGro              193\n",
       "20  MiracleGro              187\n",
       "21  MiracleGro              176"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "display(apples)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9f37ee4a",
   "metadata": {},
   "source": [
    "We obtain some commonly used descriptive statistics using the <tt>describe()</tt> method in <tt>pandas</tt>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "3832fb73",
   "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>meanAppleWeight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>22.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>179.136364</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>12.123552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>157.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>170.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>178.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>188.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>200.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       meanAppleWeight\n",
       "count        22.000000\n",
       "mean        179.136364\n",
       "std          12.123552\n",
       "min         157.000000\n",
       "25%         170.500000\n",
       "50%         178.500000\n",
       "75%         188.500000\n",
       "max         200.000000"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "apples.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "09028ae0",
   "metadata": {},
   "source": [
    "We need the descriptive statistics separately for each fertilizer type. \n",
    "\n",
    "We could use the separate dataframes that we created for plotting, but <tt>pandas</tt> has a handy method called <tt>groupby</tt> that will do the job:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "3b0c2aae",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"8\" halign=\"left\">meanAppleWeight</th>\n",
       "    </tr>\n",
       "    <tr>\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>Fertilizer</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>BrandX</th>\n",
       "      <td>10.0</td>\n",
       "      <td>173.100000</td>\n",
       "      <td>10.867382</td>\n",
       "      <td>157.0</td>\n",
       "      <td>165.00</td>\n",
       "      <td>172.5</td>\n",
       "      <td>181.00</td>\n",
       "      <td>191.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MiracleGro</th>\n",
       "      <td>12.0</td>\n",
       "      <td>184.166667</td>\n",
       "      <td>11.101460</td>\n",
       "      <td>164.0</td>\n",
       "      <td>176.75</td>\n",
       "      <td>185.5</td>\n",
       "      <td>192.25</td>\n",
       "      <td>200.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           meanAppleWeight                                               \\\n",
       "                     count        mean        std    min     25%    50%   \n",
       "Fertilizer                                                                \n",
       "BrandX                10.0  173.100000  10.867382  157.0  165.00  172.5   \n",
       "MiracleGro            12.0  184.166667  11.101460  164.0  176.75  185.5   \n",
       "\n",
       "                           \n",
       "               75%    max  \n",
       "Fertilizer                 \n",
       "BrandX      181.00  191.0  \n",
       "MiracleGro  192.25  200.0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "apples.groupby([\"Fertilizer\"]).describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1c672fd",
   "metadata": {},
   "source": [
    "It does look like the mean weight of apples from the MiracleGro trees is higher, but is the difference statistically significant?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "255f692a",
   "metadata": {},
   "source": [
    "### Carry out the test\n",
    "\n",
    "We carry out an independent samples t-test using the function <tt>ttest_ind</tt> from <tt>scipy.stats</tt>, here loaded as <tt>stats</tt>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "744b1ab0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ttest_indResult(statistic=2.350347501385599, pvalue=0.014564862730138283)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stats.ttest_ind(apples_MiracleGro[\"meanAppleWeight\"], apples_BrandX[\"meanAppleWeight\"], alternative='greater')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4dd4feae",
   "metadata": {},
   "source": [
    "The inputs to stats.ttest are the two samples to be compared (the values in the <tt>meanAppleWeight</tt> column from our separated Pandas data frames <tt>apples_BrandX</tt> and <tt>apples_MiracleGro</tt>) \n",
    "and the argument <tt>alternative='greater'</tt>, which tells the computer to run a one tailed test \n",
    "that mean of the first input <tt>apples.MiracleGro</tt> is greater than the second <tt>apples.BrandX</tt>.\n",
    "\n",
    "The outputs are <tt>statistic</tt> ($t=2.35$) and <tt>pvalue</tt> ($p=0.0146$) - if this is less than our $\\alpha$ value 0.5, there is a significant difference.\n",
    "\n",
    "### Degrees of freedom\n",
    "\n",
    "In a scientific write-up we also need to report the degrees of freedom of the test. This tells us how many observations (data-points) the test was based on, corrected for the number of means we had to estimate from the data in order to do the test.\n",
    "\n",
    "In the case of the independent samples t-test $df = n_1 + n_2 - 2$ so in this case, df=(10+12-2)=20 and we can report out test results as:\n",
    "\n",
    "$t(20) = 2.35, p=0.0146$ (one-tailed)\n",
    "\n",
    "### Interpretation\n",
    "\n",
    "Our t value of 2.35 means that the difference in mean apple weights between MiracleGro and BrandX trees is 2.35 times the standard error (where $ SE = s \\sqrt{\\frac{1}{n_1}+\\frac{1}{n_2}}$).\n",
    "\n",
    "Such a large difference (in the expected direction) would occur 0.0146 (1.46%) of the time due to chance if the null hypothesis were true (if MiracleGro was really no better than Brand X), hence the p value of 0.0146.\n",
    "\n",
    "This diagram shows the expected distribution of t-values if the null were true, with our obtained t-value marked:\n",
    "\n",
    "<img src=\"https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook/main/images/ttestApples.png\" width=50% alt=\"There should be a picture of the t-distribution here\" >"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5f263742",
   "metadata": {},
   "source": [
    "### Draw conclusions\n",
    "\n",
    "As p<0.05 we conclude that the mean weight of apples on trees fertilized with MiracleGro is indeed greater than the mean weight of apples on trees fertilized with Brand X"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99f1c539",
   "metadata": {},
   "source": [
    "## Write-up \n",
    "<br>\n",
    "\n",
    "<div style = \"    padding-top: 10px;\n",
    "    padding-bottom: 10px;\n",
    "    padding-left: 10px;\n",
    "    padding-right: 10px;\n",
    "    box-shadow: 0px 8px 16px 0px rgba(0,0,0,0.2);\n",
    "    vertical-align: middle;\">\n",
    "    \n",
    "Above, I walked you through how to run the t-test and why we make different choices. \n",
    "    \n",
    "In this section we revisit the analysis, but here we practice writing up our analysis in the correct style for a scientific report. \n",
    "    \n",
    "Replace the XXXs with the correct values!    \n",
    "\n",
    "</div>\n",
    "\n",
    "<br>"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6f5439f3",
   "metadata": {},
   "source": [
    "    \n",
    "We tested the hypothesis that the mean weight of apples produced by trees fertilized with MiracleGro was higher than for trees fertilized with Brand X.\n",
    "\n",
    "The mean apple weight was measured for each of XX trees fertilized with MiracleGro (mean of mean apple weights over XX trees XXX.Xg, sd of mean apple weights over 100 trees, XXX.Xg) and XX trees fertilized with Brand X (mean over XX trees XXX.Xg, sd XXX.Xg).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "5403735d",
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"8\" halign=\"left\">meanAppleWeight</th>\n",
       "    </tr>\n",
       "    <tr>\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>Fertilizer</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>BrandX</th>\n",
       "      <td>10.0</td>\n",
       "      <td>173.100000</td>\n",
       "      <td>10.867382</td>\n",
       "      <td>157.0</td>\n",
       "      <td>165.00</td>\n",
       "      <td>172.5</td>\n",
       "      <td>181.00</td>\n",
       "      <td>191.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MiracleGro</th>\n",
       "      <td>12.0</td>\n",
       "      <td>184.166667</td>\n",
       "      <td>11.101460</td>\n",
       "      <td>164.0</td>\n",
       "      <td>176.75</td>\n",
       "      <td>185.5</td>\n",
       "      <td>192.25</td>\n",
       "      <td>200.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           meanAppleWeight                                               \\\n",
       "                     count        mean        std    min     25%    50%   \n",
       "Fertilizer                                                                \n",
       "BrandX                10.0  173.100000  10.867382  157.0  165.00  172.5   \n",
       "MiracleGro            12.0  184.166667  11.101460  164.0  176.75  185.5   \n",
       "\n",
       "                           \n",
       "               75%    max  \n",
       "Fertilizer                 \n",
       "BrandX      181.00  191.0  \n",
       "MiracleGro  192.25  200.0  "
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "apples = pandas.read_csv('https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook/main/data/AppleWeights.csv')\n",
    "apples.groupby([\"Fertilizer\"]).describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "59fd722c",
   "metadata": {},
   "source": [
    "Theoretical considerations suggest that data for each group of trees should be drawn from a normal distribution: as individual data points were themselves the means of large samples (all apples from a given tree), these data points should follow a normal distribubtion due to the Central Limit Theorem. This was supported by a plot of the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "e06d0871",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# let's make separate dataframes for the two brands of fertilizer\n",
    "apples_BrandX = apples[apples[\"Fertilizer\"]==\"BrandX\"]\n",
    "apples_MiracleGro = apples[apples[\"Fertilizer\"]==\"MiracleGro\"]\n",
    "\n",
    "sns.kdeplot(apples_BrandX[\"meanAppleWeight\"], color='b', shade=True, label='BrandX')\n",
    "sns.kdeplot(apples_MiracleGro[\"meanAppleWeight\"], color='r', shade=True, label='MiracleGro')\n",
    "\n",
    "sns.rugplot(apples_BrandX[\"meanAppleWeight\"], color='b')\n",
    "sns.rugplot(apples_MiracleGro[\"meanAppleWeight\"], color='r')\n",
    "\n",
    "plt.xlabel(\"Mean weight of apples (g)\", fontsize = 12)\n",
    "plt.ylabel(\"Density\", fontsize = 12)\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6550a8cb",
   "metadata": {},
   "source": [
    "An independent samples t-test was therefore used to compare the means (alpha = XXX, XXX-tailed). "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "36f701e7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ttest_indResult(statistic=2.350347501385599, pvalue=0.014564862730138283)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stats.ttest_ind(apples_MiracleGro[\"meanAppleWeight\"], apples_BrandX[\"meanAppleWeight\"], alternative='greater')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e5d4dc7",
   "metadata": {},
   "source": [
    "The weight of apples from the trees fertilized with MiracleGro was indeed found to be significantly higher: t(18) = 2.35, p=0.0146."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a113d25",
   "metadata": {},
   "source": [
    "## Exercises\n",
    "\n",
    "<ol>\n",
    "<li> Can you work out how to run a two-tailed test on the data?\n",
    "    <ul>\n",
    "<li> Info on the possible values for 'alternative' can be found on the\n",
    "<a href=\"https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.ttest_ind.html\"> scipy reference page </a>\n",
    "    </ul>\n",
    "<li> What happens to the p value if you run a two tailed test instead of one tailed? Is it more or less significant? Why?\n",
    "</ol>"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}