{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Inferential statistics\n",
    "Often, we are not only interested in describing our data with descriptive statistics like the mean and standard deviation, but want to know whether two or more sets of measurements are likely to come from the same underlying distribution. We want to draw inferences from the data. This is what inferential statistics is about.\n",
    "\n",
    "To learn how to do this in python, let's use some example data:\n",
    "\n",
    "To test whether a new wonder drug increases the eye sight, Linda and Anabel ran the following experiment with student subjects:\n",
    "\n",
    "Experimental subjects were injected a saline solution containing 1nM of the wonder drug. Control subjects were injected saline without the drug. \n",
    "The drug is only effective for an hour or so. To assess the effect of the drug, eye sight was scored by testing the subjects' ability to read small text within one hour of drug injection.\n",
    "\n",
    "However, Linda and Anabel used two different experimental designs:\n",
    "1. Linda tested each student on ten consecutive days and measured the performance after each experiment. She used 50 control (saline only) and 50 experimental subjects (saline+drug) - so 100 subjects in total, 10 measurements (score after injection) per subject.\n",
    "2. Anabel only performed a single test per subject, but she measured the eye sight 30 minutes before and 30 minutes after the treatment. She tested 60 different subjects with two measurements each (score before and score after injection).\n",
    "\n",
    "Our task is now to determine whether the wonder drug really improves eye sight as tested in these two sets of experiments.\n",
    "\n",
    "Let's start with the first dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import pandas as pd\n",
    "import scipy\n",
    "\n",
    "plt.style.use('ncb.mplstyle')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>animal</th>\n",
       "      <th>treatment</th>\n",
       "      <th>score_after</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10.053951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.894092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13.447026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.579613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.482990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>12.000260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>12.277938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>13.718489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>14.272301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>13.211777</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     animal  treatment  score_after\n",
       "0         0          0    10.053951\n",
       "1         0          0     5.894092\n",
       "2         0          0    13.447026\n",
       "3         0          0     6.579613\n",
       "4         0          0     8.482990\n",
       "..      ...        ...          ...\n",
       "995      99          1    12.000260\n",
       "996      99          1    12.277938\n",
       "997      99          1    13.718489\n",
       "998      99          1    14.272301\n",
       "999      99          1    13.211777\n",
       "\n",
       "[1000 rows x 3 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# load and explore the data\n",
    "df1 = pd.read_csv('dat/5.03_inferential_stats_design1.csv')  # Linda's data\n",
    "display(df1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## What do we need to know about the data to choose the correct statistical tests?\n",
    "We want to determine whether the treatment improves of eye sight. \n",
    "What is our _Null Hypothesis_, what is our _Alternative Hypothesis_?\n",
    "\n",
    "- Null hypothesis:\n",
    "- Alternative hypothesis:\n",
    "\n",
    "What does inferential statistics do?\n",
    "With some probability:\n",
    "- prove the Null?\n",
    "- disprove the Null?\n",
    "- prove the Alternative?\n",
    "- disprove the Alternative?\n",
    "\n",
    "The p-value is a probability - probability of what?\n",
    "\n",
    "What decisions do we need to make to select the correct test?\n",
    "- normally dist?\n",
    "- equal variance?\n",
    "- signed/unsigned?\n",
    "- paired/unpaired? independence?\n",
    "\n",
    "## Plot your data!\n",
    "Okay, but what is the first thing you do when you get a bunch of data? Plot it!! Why? Can't we just look at summary statistics? No, they are not sufficient to fully describe the distribution and can be misleading!!\n",
    "\n",
    "Anscombe's quartet is a famous example that illustrates that fact. It shows 4 sets of data:\n",
    "\n",
    "![](fig/850px-Anscombe's_quartet_3.svg.png)\n",
    "\n",
    "These four data sets are very different, but they have similar statistics:\n",
    "- Mean of x: 9\n",
    "- Sample variance of x: 11\n",
    "- Mean of y: 7.50\n",
    "- Sample variance of y:\t4.125\n",
    "- Correlation between x and y:\t0.816\n",
    "- Linear regression line: y = 3.00 + 0.500x\n",
    "- Coefficient of determination of the linear regression $R^{2}$: 0.67"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "An even more extreme example - the data dinosaur:\n",
    "![](fig/DinoSequential-1.gif)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's start with analyzing the first dataset:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>animal</th>\n",
       "      <th>treatment</th>\n",
       "      <th>score_after</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10.053951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5.894092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>13.447026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6.579613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>8.482990</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>12.000260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>12.277938</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>13.718489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>14.272301</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>13.211777</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     animal  treatment  score_after\n",
       "0         0          0    10.053951\n",
       "1         0          0     5.894092\n",
       "2         0          0    13.447026\n",
       "3         0          0     6.579613\n",
       "4         0          0     8.482990\n",
       "..      ...        ...          ...\n",
       "995      99          1    12.000260\n",
       "996      99          1    12.277938\n",
       "997      99          1    13.718489\n",
       "998      99          1    14.272301\n",
       "999      99          1    13.211777\n",
       "\n",
       "[1000 rows x 3 columns]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's plot the data:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x400 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Split DataFrame into treatment and control parts\n",
    "experiment = df1[df1['treatment']==1]\n",
    "control = df1[df1['treatment']==0]\n",
    "\n",
    "# plot animal id (x-axis) vs. scores (y-axis)\n",
    "plt.figure(figsize=(12, 4))\n",
    "plt.plot(experiment['animal'], experiment['score_after'], 'or', alpha=0.2, label='Experiment')\n",
    "plt.plot(control['animal'], control['score_after'], 'ok', alpha=0.2, label='Control')\n",
    "plt.xlabel('Animal #')\n",
    "plt.ylabel('Score after [%]')\n",
    "plt.axvline(49.5, c='k')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Are all samples independent? Are they paired or unpaired?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>animal</th>\n",
       "      <th>treatment</th>\n",
       "      <th>score_after</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9.097427</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>15.234215</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>10.425896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>11.675589</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>13.972324</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>95</th>\n",
       "      <td>95</td>\n",
       "      <td>1</td>\n",
       "      <td>13.760092</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>96</th>\n",
       "      <td>96</td>\n",
       "      <td>1</td>\n",
       "      <td>13.203628</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>97</td>\n",
       "      <td>1</td>\n",
       "      <td>16.347525</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>98</td>\n",
       "      <td>1</td>\n",
       "      <td>16.062462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>99</td>\n",
       "      <td>1</td>\n",
       "      <td>13.570623</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    animal  treatment  score_after\n",
       "0        0          0     9.097427\n",
       "1        1          0    15.234215\n",
       "2        2          0    10.425896\n",
       "3        3          0    11.675589\n",
       "4        4          0    13.972324\n",
       "..     ...        ...          ...\n",
       "95      95          1    13.760092\n",
       "96      96          1    13.203628\n",
       "97      97          1    16.347525\n",
       "98      98          1    16.062462\n",
       "99      99          1    13.570623\n",
       "\n",
       "[100 rows x 3 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "N before averaging 1000 , N after averaging 100\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 600x400 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Data from design 1 - average the repeated tests for each animal\n",
    "\n",
    "dict_grouped = {'animal': [], 'treatment': [], 'score_after': []}  # initialize a dictionary with empty lists for holding the values\n",
    "for animal in df1['animal'].unique():  # iterate over all animals\n",
    "    this_animal = df1[df1['animal']==animal]  # get the part of the table for the current animal\n",
    "    dict_grouped['animal'].append(animal)  # store the animal number\n",
    "    dict_grouped['treatment'].append(this_animal['treatment'].iloc[0])  # store the treament (0/1)\n",
    "    dict_grouped['score_after'].append(np.mean(this_animal['score_after']))  # compute and store the trial-averaged score for the current animal\n",
    "\n",
    "# make DataFrame from dict\n",
    "df1_avg = pd.DataFrame(dict_grouped)\n",
    "df1_avg.to_csv('dat/5.03_inferential_stats_design1_clean.csv')\n",
    "\n",
    "display(df1_avg)\n",
    "print('N before averaging', df1.shape[0],', N after averaging', df1_avg.shape[0])\n",
    "\n",
    "# Plot the aggregated data as a control\n",
    "experiment = df1_avg[df1_avg['treatment']==1]\n",
    "control = df1_avg[df1_avg['treatment']==0]\n",
    "\n",
    "plt.figure(figsize=(6, 4))\n",
    "plt.plot(experiment['animal'], experiment['score_after'], 'or', alpha=0.2, label='Experiment')\n",
    "plt.plot(control['animal'], control['score_after'], 'ok', alpha=0.2, label='Control')\n",
    "plt.xlabel('Animal #')\n",
    "plt.ylabel('Score after [%]')\n",
    "plt.axvline(49.5, c='k')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### One-sided or two-sided?\n",
    "\n",
    "![](5.03_inferential_stats_1.png)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Is the data normally distributed?\n",
    "The t-test assumes that the data in the sample are normally distributed in the population. This means that the values within each group should follow a normal (Gaussian) distribution. This assumption is related to the parameters of the normal distribution, such as the mean and standard deviation.\n",
    "\n",
    "To check for normality, we visualize the distributions, then run a statistical test. In `scipy.stats`, there are several tests for normality. We will use `scipy.stats.normaltest`, which derives a p-values based on the skew and kurtosis of the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# histogram the data\n",
    "# Data from design 1\n",
    "experiment = df1_avg[df1_avg['treatment']==1]\n",
    "control = df1_avg[df1_avg['treatment']==0]\n",
    "\n",
    "plt.hist(control['score_after'], color='k', alpha=0.33, label='Control')\n",
    "plt.hist(experiment['score_after'], color='r', alpha=0.33, label='Experiment')\n",
    "plt.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mini exercise: Test for normality\n",
    "Do we run the test on the full dataset? Or on the individual groups (treatment and control) separately?\n",
    "\n",
    "How do we interpret the p-values? What's the null hypothesis when we test for normality?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "NormaltestResult(statistic=0.977665116203375, pvalue=0.61334201754807)\n",
      "NormaltestResult(statistic=1.3550283711604612, pvalue=0.5078779147507373)\n"
     ]
    }
   ],
   "source": [
    "# your solution here\n",
    "print(scipy.stats.normaltest(df1_avg[df1_avg['treatment']==0]['score_after']))\n",
    "print(scipy.stats.normaltest(df1_avg[df1_avg['treatment']==1]['score_after']))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If the p-value is less than the significance level, you may reject the null hypothesis and conclude that there is a statistically significant difference. This means:\n",
    "- With 95% probability, the sample does not originate from a normal distribution with the mean value μ0.\n",
    "- In 5% of cases, however, the significant difference may have been a result of chance, within the distribution of the null hypothesis."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Equal variance?\n",
    "The independent samples t-test (and its variants) assumes that the variances within the groups being compared are equal (homoscedasticity). In other words, it assumes that the spread or dispersion of the data is consistent across groups. This assumption is also related to population parameters. \n",
    "\n",
    "However, homoscedasticity is not a hard criterion, since there exitss a variant of the t-test - Welch's t-test - that accounts for unequal variance (heteroscedasticity).\n",
    "\n",
    "Homoscedasticity is typically tested visually - by inspecting the data distributions - and rarely tested in practice. There do exist tests that compare the variance across multiple groups, like Levene test (`scipy.stats.levene`).\n",
    "\n",
    "But, it's best to use statistical tests that do not require equal variance, like Welch's variant of the t-test."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Mini Exercise: Run the tests\n",
    "We now know all we need to know about our samples to select the correct test:\n",
    "- paired or unpaired: no\n",
    "- normal: yes\n",
    "- homoscedasticity: ?\n",
    "- one/two-sided: one\n",
    "\n",
    "Check the docs to figure out how to use the correct test:\n",
    "- unpaired (independent):\n",
    "    - parametric: `scipy.stats.ttest_ind` ([docs](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.ttest_ind.html))\n",
    "    - non-parametric (for non-normal data): `scipy.stats.mannwhitneyu` ([docs](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.mannwhitneyu.html))\n",
    "- paired (or related):\n",
    "    - parametric: `scipy.stats.ttest_rel` ([doc](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.ttest_rel.html))\n",
    "    - non-parametric (for non-normal data): `scipy.stats.wilcoxon` ([doc](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.wilcoxon.html))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "TtestResult(statistic=-12.363139612077164, pvalue=5.109456570496495e-22, df=97.9954921844401)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# your solution here\n",
    "control = df1_avg[df1_avg['treatment']==0]\n",
    "treatment = df1_avg[df1_avg['treatment']==1]\n",
    "\n",
    "scipy.stats.ttest_ind(control['score_after'], treatment['score_after'], equal_var=False, alternative='less')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### If you want to compare more than two groups\n",
    "\n",
    "First, detect group-level effects using \n",
    "- Anova: `scipy.stats.f_oneway`\n",
    "- non-parameteric alternative: Kruskal-Wallis test `scipy.stats.kruskal`\n",
    "\n",
    "If p<0.05, there exist a difference between the groups.\n",
    "\n",
    "To then detect which groups are different, you run a post hoc test:\n",
    "\n",
    "`scipy.stats.tukey_hsd` or `scipy.stats.dunnett`\n",
    "\n",
    "You will do this in one of the bonus exercises."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "7ea0ec616133ead53c1908c8f6539f5c0cb9b2f78368e2bb6ab3f847e89ca400"
  },
  "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": "3.12.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}