{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "f47a894a",
   "metadata": {},
   "source": [
    "# Vacation Assignment\n",
    "\n",
    "This assignment has two parts. \n",
    "\n",
    "In the first part you carry out some hypothesis tests; this is designed for you to practice both the process of running the different tests, and to make sure you understand the choices you are making when running tests.\n",
    "\n",
    "The second part, 'revision', targets some Python syntax and (mmainly) some concepts that I want to make sure everyone has securely grasped. The answers to all conceptual quesitons can be found in the lectures and reading; searching for them is good revision!\n",
    "\n",
    "Overall, the idea of this assignment is to help you consolidate what you learned this term and it should prepare you well for the collection in January. Please remember that searching for the answers (in the lecture notes etc) and trying things til it works (for coding) is part of the learning process; you are *supposed* to need to do that rather than having all the answers at your fingertips.\n",
    "\n",
    "If you get stuck on something, try the lecture notes, try the reading, and try Google as there are many helpful 'how to' blogs out there and many blogs explaining statistical concepts. If you still are stuck after all that I suggest you leave a note on the particular question part for your tutor, perhaps offering your best guess at the answer. Try not to email them as they need a break over the vacation!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a47712cd",
   "metadata": {
    "tags": []
   },
   "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 as pd\n",
    "import seaborn as sns\n",
    "sns.set_theme(style='white')\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.formula.api as smf"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "849cc000-0637-4bc3-8572-294947012ceb",
   "metadata": {
    "tags": []
   },
   "source": [
    "# Part 1 - data analysis exercises\n",
    "\n",
    "In these questions you need to carry out various statistical tests. Please read the questions carefully and answer all parts - in particular if you are asked to *comment* on your results you must comment; this usually means interpret the results in plain English."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ece8bc7",
   "metadata": {},
   "source": [
    "## Question 1 - Cloud seeding\n",
    "\n",
    "Cloud seeding is a process by which a light aircraft dumps particulate matter into a cloud in the hope of causing precipitation (rain). The file CloudSeeding.csv contains standardized rainfall yield measures for seeded and unseeded clouds. These are real data from a study in 1975.\n",
    "\n",
    "As a side note, cloud seeding does work and <a href='https://www.independent.co.uk/sport/olympics/how-beijing-used-rockets-to-keep-opening-ceremony-dry-890294.html'> is used to *prevent* rain </a> during important outdoor events - although it mimght have the side effect of causing a deluge nearby!\n",
    "\n",
    "**a)  Download the data file and load the data into a `Pandas` dataframe called `clouds`**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "1cd63096",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "clouds = pd.read_csv('https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook_2024/main/data/cloudSeeding.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80e909ec",
   "metadata": {},
   "source": [
    "**b) Plot the data for seeded and unseeded clouds in a way that shows the distribution of rainfall yields**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "c94b7bb7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "35d406d4",
   "metadata": {},
   "source": [
    "**c) Calculate the parameters of the best fitting Normal distributions for seeded and unseeded clouds respectively**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a16ff627-cf7d-44a9-9ca3-85363cbbbb9e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bc4a84d2-151e-424b-a5c4-a5d1f03c2571",
   "metadata": {},
   "source": [
    "<font color='red'>State the parameters here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f81fed9b-44a0-4f9d-a882-8b449452df08",
   "metadata": {},
   "source": [
    "**d) Is a normal distribution a good fit to the data?**\n",
    "\n",
    "*The here we overlay the best fitting normal on the data KDE plot:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "b9f7c1c5",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x500 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 5))\n",
    "\n",
    "seeded=clouds.query('status==\"Seeded\"').rainfall\n",
    "unseeded=clouds.query('status==\"Unseeded\"').rainfall\n",
    "\n",
    "# seeded clouds\n",
    "plt.subplot(2,1,1)\n",
    "x=range(-1000,4000)\n",
    "y = stats.norm.pdf(x,seeded.mean(),seeded.std())\n",
    "plt.plot(x,y,color=(1,0,0))\n",
    "sns.kdeplot(data=seeded, color=[1,0,0], fill=True)\n",
    "sns.rugplot(data=seeded, color=[1,0,0], height=0.1)\n",
    "plt.ylim([0,0.0022])\n",
    "\n",
    "# unseeded clouds\n",
    "plt.subplot(2,1,2)\n",
    "x=range(-1000,4000)\n",
    "y = stats.norm.pdf(x,unseeded.mean(),unseeded.std())\n",
    "plt.plot(x,y,color=(0,0,1))\n",
    "sns.kdeplot(data=unseeded, color=[0,0,1], fill=True)\n",
    "sns.rugplot(data=unseeded, color=[0,0,1], height=0.1)\n",
    "plt.ylim([0,0.0022])\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b57d260b",
   "metadata": {},
   "source": [
    "*Based on the plot above, the data are not well fitted by a normal distribution. Comment on the what you can see in the graph that suggests the normal distribution is a poor fit.*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4654f9d5-d736-413d-8b83-b1f288470686",
   "metadata": {},
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "19f0ddd6",
   "metadata": {},
   "source": [
    "**e) Conduct three statistical tests of the researcher’s hypothesis, based on:**\n",
    "\n",
    "*I.\tAssumption of normality*\n",
    "\n",
    "*II.\tPermutation*\n",
    "\n",
    "*III.\tRanked values*\n",
    "\n",
    "*… in each case you should:*\n",
    "* State the null and alternative hypothesis and the alpha value\n",
    "* Report appropriate descriptive statistics for the test in question\n",
    "* Carry out the test\n",
    "* State the results of the test including the test statistic and p value. For the t-test, degrees of freedom should also be stated.\n",
    "* Report your practical conclusion in plain English\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "62ab3dad",
   "metadata": {},
   "source": [
    "**f) Comment on the relative merits of each possible choice of test for this dataset.**\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "250de123",
   "metadata": {},
   "source": [
    "## Question 2: Colouring books\n",
    "\n",
    "A researcher hypothesises that working on adult colouring books is particularly relaxing and will lower resting heart rate.\n",
    "\n",
    "She measures resting heart rate after one hour of colouring and after one hour of reading a novel in the same participants. The data are provided in the file ColouringHeartRate.csv. These are made-up data.\n",
    "\n",
    "Download the data file and load the data into a Pandas dataframe called heartRate:\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "cc692ae9",
   "metadata": {
    "tags": []
   },
   "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>participantID</th>\n",
       "      <th>colouring</th>\n",
       "      <th>reading</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>a</td>\n",
       "      <td>62</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b</td>\n",
       "      <td>66</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>c</td>\n",
       "      <td>61</td>\n",
       "      <td>69</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>d</td>\n",
       "      <td>68</td>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>e</td>\n",
       "      <td>74</td>\n",
       "      <td>75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>f</td>\n",
       "      <td>62</td>\n",
       "      <td>71</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>g</td>\n",
       "      <td>68</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>h</td>\n",
       "      <td>62</td>\n",
       "      <td>72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>i</td>\n",
       "      <td>55</td>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>j</td>\n",
       "      <td>59</td>\n",
       "      <td>65</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  participantID  colouring  reading\n",
       "0             a         62       71\n",
       "1             b         66       71\n",
       "2             c         61       69\n",
       "3             d         68       61\n",
       "4             e         74       75\n",
       "5             f         62       71\n",
       "6             g         68       77\n",
       "7             h         62       72\n",
       "8             i         55       62\n",
       "9             j         59       65"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "heartRate = pd.read_csv('https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook_2024/main/data/ColouringHeartRate.csv')\n",
    "heartRate"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "eabd1e63",
   "metadata": {},
   "source": [
    "**b) Plot the data and comment.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "fd44ebd0",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60c8db07",
   "metadata": {},
   "source": [
    "<font color='red'> Your comments here!"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "296def8a",
   "metadata": {},
   "source": [
    "**c)\tThis is a within-subjects or repeated-measures design. With reference to your graph, explain the advantages of a within-subjects design in this particular experiment.**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c67b2ae9",
   "metadata": {},
   "source": [
    "**d) Control condition**\n",
    "\n",
    "* Why do you think the researcher chose to compare heart rate after one hour of colouring to heart rate after one hour of reading, as opposed to comparing heart rate before and after colouring?*\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bf0438cf-5f97-49fd-8a1e-1512fb555ed8",
   "metadata": {},
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f3c2e095",
   "metadata": {},
   "source": [
    "**e)\tConduct three statistical tests of the researcher’s hypothesis, based on** \n",
    "\n",
    "*I.\tAssumption of normality*\n",
    "\n",
    "*II.\tPermutation*\n",
    "\n",
    "*III.\tRanked values*\n",
    "\n",
    "*… in each case you should:*\n",
    "\n",
    "* State the null and alternative hypothesis and the alpha value\n",
    "* Report appropriate descriptive statistics for the test in question\n",
    "* Carry out the test\n",
    "* State the results of the test including the test statistic and p value. For the t-test, degrees of freedom should also be stated.\n",
    "* Report your practical conclusion in plain English\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48529841-eb6f-4429-91f1-74288fcf8965",
   "metadata": {},
   "source": [
    "**f) Assumption of normality**\n",
    "\n",
    "*The use of the t-test rests on an assumption of normality. For a paired test, it is the **differences** that must be normally distributed for the t-test to be valid*\n",
    "\n",
    "*Can you explain why it is the distribution of the differences, rather than the data, that is important?*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "91189c7c-6aef-4e10-b493-155c298b22f1",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "757450ed",
   "metadata": {},
   "source": [
    "*Let's check if the distribution of differences is normal:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "4f8b44e7",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "heartRate['diff']=heartRate.colouring-heartRate.reading\n",
    "sns.kdeplot(heartRate['diff'], fill=True)\n",
    "sns.rugplot(heartRate['diff'], height=0.1)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "554d2e58-567e-4a9d-bc55-810dbce8df7f",
   "metadata": {},
   "source": [
    "*It seems not - the differences have a distribution with positive skew (to be fair the sample size is too small to be really confident about this, but if unsure, it is better to avoid assuming normality)*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "01f3bb51",
   "metadata": {},
   "source": [
    "**g) Which statistical test do you think was the best choice for this dataset?**\n",
    "\n",
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f67ea49a",
   "metadata": {},
   "source": [
    "## Question 3: Reaction times\n",
    "\n",
    "*Two participants, A and B, perform a choice reaction time task in which they must press one of two buttons depending on whether a word was a real or made-up word*\n",
    "\n",
    "*The data are provided in the file ChoiceRTs.csv. These are made-up data*\n",
    "\n",
    "*Download the data file and load the data into a Pandas dataframe called ChoiceRTs:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "991e8f15",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "ChoiceRTs = pd.read_csv('https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook_2024/main/data/ChoiceRTs.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0ec54b7",
   "metadata": {},
   "source": [
    "**a) Plot the data**\n",
    "\n",
    "*Plot the data for each participant in a way that illustrates the distribution of reaction times for each person and allows them to be compared. Two subplots within a single figure would be a good choice here.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "80332622",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b568043a",
   "metadata": {},
   "source": [
    "**c) Comment on the plot**\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "92cb7241-efd0-4eb3-a47e-1f8fc542d225",
   "metadata": {},
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "934b2b15",
   "metadata": {},
   "source": [
    "**d) Data cleaning decision**\n",
    "\n",
    "*The researcher decides that reaction times under 300ms (very fast responses) and over 700ms (very slow responses) should be excluded.*\n",
    "\n",
    "*Do you think this is justified? Explain your answer.*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "25d6531b-462b-4620-8346-309d5d4bfaee",
   "metadata": {},
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5be38f0e-3a2b-4002-9312-cd448733ff13",
   "metadata": {
    "tags": []
   },
   "source": [
    "**e) Data cleaning implementation**\n",
    "\n",
    "*Make a new dataframe called `ChoiceRTs_clean` with the data with RTs outside the range 300-700ms replaced by `NaN`*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "d53a1374",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c69c2fd8-5faa-478f-aaa1-d81788410afa",
   "metadata": {},
   "source": [
    "**f) Calculate the mean and sd of reaction time**\n",
    "\n",
    "* before and \n",
    "* after excluding data outside the range 300-700 ms\n",
    "\n",
    "*Comment on the difference in results*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "99b60bdb",
   "metadata": {},
   "source": [
    "**g) Statistical test**\n",
    "\n",
    "*In each case:*\n",
    "\n",
    "i)\tall the data\n",
    "\n",
    "ii) the data with RTs outside the range 300-700ms excluded\n",
    "\n",
    "*conduct a permutation test to determine whether there is a difference in the mean reaction time between the two participants*\n",
    "\n",
    "**HINT you will need to replace the function `np.mean()` with `np.nanmean()` as np.mean() returns a NaN if any of the values being averaged are NaN**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "670a1d6f",
   "metadata": {},
   "source": [
    "**h) Comment on the difference in results when outliers are excluded**\n",
    "\n",
    "Comment on the difference in results for the permutation test, with data outside the range 300-700ms omitted or not\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "811a8459-c0d4-48f4-91b9-57d500a65054",
   "metadata": {},
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1df700fe-5099-46a1-862d-abf3d9a74aac",
   "metadata": {
    "tags": []
   },
   "source": [
    "## Part 2 - More Revision\n",
    "\n",
    "### Long vs wideform data\n",
    "\n",
    "*Pandas dataframes can be in longform or wideform (definitions <a href='https://www.statology.org/long-vs-wide-data/'>here</a>. We have seen examples of both throughout the term. Let's check you are comfortable with using both formats.*\n",
    "\n",
    "*Here are same data are provided in longform and wideform:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "c1d3de6e-8cd4-4a48-bc98-a597339fda7d",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "heightsWideform = pd.read_csv('https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook_2024/main/data/BrotherSisterData.csv')\n",
    "heightsLongform = pd.read_csv('https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook_2024/main/data/BrotherSisterDataLongform.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b80c0d6a-663f-4ca5-a677-13497fc4ec3b",
   "metadata": {},
   "source": [
    "**a. Plot the data as a KDE plot (plot brothers and sisters as two separate KDEs)**\n",
    "\n",
    "You might need to look back through the notes for examples of how to do this, or just try a few options to see what works."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "21454aa1-063d-405b-b7d1-5fc03800dd04",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here to plot the longform data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "42f4a463-27e0-44bf-acd2-62c7c210f9f7",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here to plot the wideform data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "78bcd8e7-85e2-49cd-91ae-44a72332e0df",
   "metadata": {
    "tags": []
   },
   "source": [
    "**b. Plot the data as a scatterplot of brother vs sister's height**\n",
    "\n",
    "*Let's do this for both longform and wideform data. **Use the function `sns.regplot()`** which adds the best fitting regression line (the straight line that best fits through the data points) and add the line x=y:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "8cfe1d8a-7e32-42f6-99b6-c3aa2315bcfe",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here to plot the wideform data\n",
    "# use sns.regplot to make a sctterplot with regression line\n",
    "\n",
    "# add the line x=y as a red dashed line\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "489431e9-e9b1-44ae-a090-69db8d222147",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here to plot the longform data\n",
    "# use sns.regplot to make a sctterplot with regression line\n",
    "\n",
    "# add the line x=y as a red dashed line\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "db448951-5143-4a85-8eca-bcb337b71e74",
   "metadata": {},
   "source": [
    "**c. Get the mean height for brothers and sisters separately**\n",
    "\n",
    "*Do this for the longform and wideform dataframes*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "b9962fa9-f34e-421f-b2a2-76658d6f9cad",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code to get the mean brother and sister heights from the wideform data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "5f39635f-227f-4b0a-9d34-03782e13ec3e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code to get the mean brother and sister heights from the wideform data"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "dcbfdf81-0ae9-42d9-81f7-7a34a5192a2c",
   "metadata": {
    "tags": []
   },
   "source": [
    "### Experimental design\n",
    "\n",
    "**a. Paired design**\n",
    "\n",
    "*i) What is a *paired* design (also called a *matched-pairs* design)?*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b98dd1b8-eac2-42e0-97ca-bc2bdb1912d3",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e12f73e5-1a8f-4d97-93a2-09250ade00a4",
   "metadata": {
    "tags": []
   },
   "source": [
    "*ii) What are the advantages of a paired design (compared to a independent samples design?)*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "257c46b4-a286-4f49-96de-42110be0975c",
   "metadata": {},
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "21be6e56-bf6f-4e6a-b102-f0b75d92f4e4",
   "metadata": {},
   "source": [
    "*iii) In a paired design, we are testing whether the average *difference* withing pairs is significantly different from zero. For example, is the difference between each brother and his own sister reliably positive (brothers are taller than sisters)?*\n",
    "\n",
    "*Look at the scatterplot above. In this case, why is it particularly helpful to use a* paired *design rather than an independent samples design?*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ffbfcd97-abce-46dc-a85d-ce1be778d9f6",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "082b7e92-614e-436c-bbc3-a536e2f6ac8d",
   "metadata": {},
   "source": [
    "**b) Repeated measures design**\n",
    "\n",
    "*A **repeated measures** design is a type of paired design.*\n",
    "\n",
    "*i) Explain what is the difference between a paired design (in general) and a repeated measures design.*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b638190a-d59c-49b7-a519-2e7118c7d693",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "acc46cfc-56c6-4ae2-90d0-191d872712ad",
   "metadata": {},
   "source": [
    "*ii) Give an example of a study in which a repeated measures design could be used (ie, 'if we wanted to test xxx, we could use a repeated measures design in which we measure.....')*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "157379f0-2f94-459d-97c0-6d257efdabfc",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "82702ddf-d6b7-4715-a8b9-b5237bc86583",
   "metadata": {},
   "source": [
    "*iii) What potential pitfalls are there with repeated measures designs? How might they be mitigated?*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9e043628-1590-4864-a4ec-d669d4b15460",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7dcde944-97ad-4656-947c-d3a96cab8c0a",
   "metadata": {},
   "source": [
    "**d) Detailed example**\n",
    "\n",
    "*A researcher hypothesises that men are taller than women:*\n",
    "\n",
    "$\\mathcal{H_o}$: the mean height of men is equal to that of women\n",
    "\n",
    "$\\mathcal{H_a}$: the mean height of men is greater than that of women\n",
    "\n",
    "*She conducts two permutation tests on the brother-sister data as follows:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "f1a3abc5-5e9a-4a5f-8380-aee4dadd492c",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PermutationTestResult(statistic=4.640000000000015, pvalue=0.0784, null_distribution=array([-0.16,  2.48,  4.08, ...,  4.88,  2.8 ,  3.28]))"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# TEST 1:\n",
    "def dMeans(x,y):\n",
    "    return np.mean(x)-np.mean(y)\n",
    "\n",
    "results1 = stats.permutation_test((heightsWideform.brother, heightsWideform.sister), \n",
    "                                   dMeans, permutation_type='independent')\n",
    "results1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "7ed54604-15bb-4d7a-9c93-e556310bd0c7",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PermutationTestResult(statistic=4.640000000000015, pvalue=0.0002, null_distribution=array([ 0.32,  0.  , -0.96, ...,  0.32, -0.4 ,  0.48]))"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# TEST 2:\n",
    "def mDiff(x,y):\n",
    "    return np.mean(x-y)\n",
    "\n",
    "results2 = stats.permutation_test((heightsWideform.brother, heightsWideform.sister), \n",
    "                                   dMeans, permutation_type='samples')\n",
    "results2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3366bc95-4a59-4311-86e1-a275019acedb",
   "metadata": {},
   "source": [
    "**i) Which test was correct and why?**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e56b6700-a0b7-4896-b229-6f2bf9f9bbe9",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "56eb63f2-4278-4593-943e-fa4e0ef8bdfc",
   "metadata": {},
   "source": [
    "**ii) The p-value for Test 2 is smaller, why?**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c0f673b-8beb-4fbc-9abb-60b4ec75e152",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "81199a5d-c566-476a-ad8a-7ebe8c86823f",
   "metadata": {},
   "source": [
    "**iii) In each case we specify a function to get the difference of means - what is the difference between `dMeans()` and `mDiff()`?**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0f9ea569-9efa-42e8-88ed-e1722f7f9278",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here\n",
    "    "
   ]
  },
  {
   "cell_type": "markdown",
   "id": "072dbf3f-4237-4466-befe-7d046b4fb437",
   "metadata": {},
   "source": [
    "**iv) The Test Statistic is the same for tests 1 and 2. Can you explain why?**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0e04c0f6-5d13-4bcd-9968-8f8e94e10d97",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "819dea11-3df6-4a05-b97e-8c1e95fedb8a",
   "metadata": {
    "tags": []
   },
   "source": [
    "**v) The null distribution for tests 1 and 2 is quite different. Can you explain why?**\n",
    "\n",
    "The null distributions are plotted for you here to help:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "ffc953a8-6c8e-4a1e-8d1e-aecb829c5d53",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.subplot(1,2,1)\n",
    "sns.histplot(results1.null_distribution)\n",
    "plt.xlim([-10,10])\n",
    "plt.xlabel('difference of group means')\n",
    "\n",
    "plt.subplot(1,2,2)\n",
    "sns.histplot(results2.null_distribution)\n",
    "plt.xlim([-10,10])\n",
    "plt.xlabel('mean pairwise difference')\n",
    "\n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b664a3a3-5abc-45b0-9fc7-db94b0516ede",
   "metadata": {},
   "source": [
    "<font color='red'>Your answer here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a4d37301-7e6f-4501-b86a-3377b0616f5f",
   "metadata": {
    "tags": []
   },
   "source": [
    "### One vs Two-tailed tests\n",
    "\n",
    "*If we know the direction in which we expect an effect (men are taller than women) we can run a one-tailed test. Otherwise we must run a two-tailed test.*\n",
    "\n",
    "*Load the following (made-up) data comparing the heights of female Psychology and female BMS students:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "cb51ff87-47be-414d-800f-ba8744824258",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "heights = pd.read_csv('https://raw.githubusercontent.com/jillxoreilly/StatsCourseBook_2024/main/data/PsyBMSheights.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1f99cbe4-8a8b-466b-8b65-0fa4ed2e8c8f",
   "metadata": {},
   "source": [
    "*A researcher hypothesises that BMS students are taller than psychology students. She limits her analysis to female students:*\n",
    "    \n",
    "$\\mathcal{H_o}$ The mean height of (female) psychology and BMS students are the same\n",
    "\n",
    "$\\mathcal{H_a}$ The mean height of (female) psychology students is less than the mean height of BMS students\n",
    "\n",
    "*She decides to test at the $\\alpha=0.05$ level*\n",
    "\n",
    "**a) Is this a one or two tailed test?**"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "17f8ef9d-a411-47cf-bf09-8f959bc83b09",
   "metadata": {},
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "33af1492-01e3-4311-8a4f-62b871dc9fd5",
   "metadata": {},
   "source": [
    "**b) What's wrong in the following example?**\n",
    "\n",
    "The researcher used a t-test to test her hypothesis as follows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "206572a3-8e86-43e9-b8ee-ce93dd0b00be",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Ttest_indResult(statistic=1.823852779167176, pvalue=0.9627982812018988)"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stats.ttest_ind(heights.query('subject==\"BMS\"').height, heights.query('subject==\"Psychology\"').height, alternative='less')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "336f73b3-a65a-4fdd-83d0-b2dfb20904c9",
   "metadata": {},
   "source": [
    "What has she done wrong?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "27a8b901-bc8e-4711-83cf-bb61ad36df14",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3a2c1841-69de-415d-aedb-5db4128901ed",
   "metadata": {},
   "source": [
    "**c) Correct the mistake**\n",
    "\n",
    "*Re-run the test, correcting the researcher's mistake*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "37c8a571-a542-4cba-b3a4-b21a7768eafe",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "503478d1-35d8-479f-9659-4ada7c3e2d0b",
   "metadata": {},
   "source": [
    "**d) Run a two-tailed test**\n",
    "\n",
    "*Say the researcher had no a-priori hypothesis about which subject hahs taller students. Then she should have stated her hypotheses as follows, and run a two-tailed test*\n",
    "\n",
    "\n",
    "$\\mathcal{H_o}$ The mean height of (female) psychology and BMS students are the same\n",
    "\n",
    "$\\mathcal{H_o}$ The mean height of (female) psychology students is different from mean height of BMS students\n",
    "\n",
    "*She still wishes to test at the $\\alpha=0.05$ level**\n",
    "\n",
    "**Run the test yourself**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "c255bd49-a20b-46dd-8e9c-64de96f7580e",
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "acb1e4da-ccaa-4f2a-b002-02ac94743938",
   "metadata": {},
   "source": [
    "*i) How does the $p$-value from the two tailed test relate to that of (correct version of) the one-tailed test above?*"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ab242e0c-d2d5-4ece-b830-3f3adf24c3ec",
   "metadata": {},
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ad6e1659-fd06-437d-9e8f-1232c9cd3e19",
   "metadata": {},
   "source": [
    "*ii) The two-tailed test is not significant at the 5% level whilst the one-tailed test was significant. Can you explain why?*\n",
    "* Key terms for your answer are the *critical region* of the test and the *critical value* or *cut-off value*\n",
    "* These were introduced in lectures but if you don't remember them, a quick good brings up several tutorials"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0c1cc683-830c-4007-9433-9ea6acd51b66",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4e877c66-32e4-41ca-a180-6a13afa6fe06",
   "metadata": {},
   "source": [
    "**e) \"Peeking\"**\n",
    "\n",
    "*Why did the researcher think that BMS students would be taller in the first place? Because when she plotted the data to check for outliers (which is indeed correct procedure), she noticed that the BMS students looked taller:*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "06590f82-1a49-40dd-8dbd-e831995d3798",
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.kdeplot(data=heights, x='height', hue='subject', fill=True)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "80ed19ce-9392-492b-b9d0-0c34b2a45b29",
   "metadata": {},
   "source": [
    "If the psychology students had looked taller, she would have done a one-tailed test in the other direction.\n",
    "\n",
    "Using this approach (sometiems called \"peeking\" at the data) is wrong, because means that the chance of getting a 'significant' result if the null was true (the false positive rate) is double.\n",
    "\n",
    "Can you explain why?"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7e3ef6bb-74f1-41da-858d-218f7319ca84",
   "metadata": {
    "tags": []
   },
   "source": [
    "<font color='red'>Your answer here"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ef22b606-1c9b-46ac-8034-35586082bdb3",
   "metadata": {},
   "source": [
    "*It is not wrong to inspect your data before running a test, but **your hypotheses (and in particular justification for a one-tailed test) should reflect predictions you would be able to make without seeing the particular sample of data**.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "2eadb72a-af9c-4bcd-9537-9c4099cc7219",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "54ca3aa6-edf0-4586-b2c9-7699415eb9b5",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.9.18"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}