{
  "nbformat": 4,
  "nbformat_minor": 5,
  "metadata": {
    "kernelspec": {
      "display_name": "Python 3",
      "language": "python",
      "name": "python3"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.8"
    },
    "colab": {
      "provenance": []
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6yDxo9u0d9Na"
      },
      "source": [
        "#**Project Business Statistics: E-news Express**\n",
        "\n",
        "- Student: Alexey Tyurin\n",
        "- Group: Oct'22 C Sun - MLS Grp B\n",
        "- Batch: PGP-DSBA-UTA-OCT22-C\n",
        "- Date: 1/25/2023"
      ],
      "id": "6yDxo9u0d9Na"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SrLd-ICEmPSD"
      },
      "source": [
        "##**Define Problem Statement and Objectives**"
      ],
      "id": "SrLd-ICEmPSD"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9Vt1PLLqmXNn"
      },
      "source": [
        "###**Business Context**\n",
        "The advent of e-news, or electronic news, portals has offered us a great opportunity to quickly get updates on the day-to-day events occurring globally. The information on these portals is retrieved electronically from online databases, processed using a variety of software, and then transmitted to the users. There are multiple advantages of transmitting new electronically, like faster access to the content and the ability to utilize different technologies such as audio, graphics, video, and other interactive elements that are either not being used or aren’t common yet in traditional newspapers.\n",
        "\n",
        "E-news Express, an online news portal, aims to expand its business by acquiring new subscribers. With every visitor to the website taking certain actions based on their interest, the company plans to analyze these actions to understand user interests and determine how to drive better engagement. The executives at E-news Express are of the opinion that there has been a decline in new monthly subscribers compared to the past year because the current webpage is not designed well enough in terms of the outline & recommended content to keep customers engaged long enough to make a decision to subscribe.\n",
        "\n",
        "[Companies often analyze user responses to two variants of a product to decide which of the two variants is more effective. This experimental technique, known as A/B testing, is used to determine whether a new feature attracts users based on a chosen metric.]"
      ],
      "id": "9Vt1PLLqmXNn"
    },
    {
      "cell_type": "markdown",
      "source": [
        "###**Objective**\n",
        "The design team of the company has researched and created a new landing page that has a new outline & more relevant content shown compared to the old page. In order to test the effectiveness of the new landing page in gathering new subscribers, the Data Science team conducted an experiment by randomly selecting 100 users and dividing them equally into two groups. The existing landing page was served to the first group (control group) and the new landing page to the second group (treatment group). Data regarding the interaction of users in both groups with the two versions of the landing page was collected. Being a data scientist in E-news Express, you have been asked to explore the data and perform a statistical analysis (at a significance level of 5%) to determine the effectiveness of the new landing page in gathering new subscribers for the news portal by answering the following questions:\n",
        "\n",
        "1. Do the users spend more time on the new landing page than on the existing landing page?\n",
        "2. Is the conversion rate (the proportion of users who visit the landing page and get converted) for the new page greater than the conversion rate for the old page?\n",
        "3. Does the converted status depend on the preferred language?\n",
        "4. Is the time spent on the new page the same for the different language users?"
      ],
      "metadata": {
        "id": "rm_8aJoVW73-"
      },
      "id": "rm_8aJoVW73-"
    },
    {
      "cell_type": "markdown",
      "source": [
        "###**Data Dictionary**\n",
        "The data contains information regarding the interaction of users in both groups with the two versions of the landing page.\n",
        "\n",
        "1. `user_id` - Unique user ID of the person visiting the website\n",
        "2. `group` - Whether the user belongs to the first group (control) or the second group (treatment)\n",
        "3. `landing_page` - Whether the landing page is new or old\n",
        "4. `time_spent_on_the_page` - Time (in minutes) spent by the user on the landing page\n",
        "5. `converted` - Whether the user gets converted to a subscriber of the news portal or not\n",
        "6. `language_preferred` - Language chosen by the user to view the landing page"
      ],
      "metadata": {
        "id": "csG6d7DiW_t9"
      },
      "id": "csG6d7DiW_t9"
    },
    {
      "cell_type": "markdown",
      "source": [
        "###**Solution Approach**\n",
        "\n",
        "To solve the above problem, we need to answer four questions:\n",
        "1. Is the average time users spend on the new landing page significantly greater than the average time spent on the existing landing page?\n",
        "2. Is the difference in proportions of the users who visited the landing pages (existing and new) and got converted (conversion rate) for the new page significantly greater to conclude that the conversion rate for the new page is greater than the conversion rate for the existing page?\n",
        "3. Is the conversion and preferred language independent or related?\n",
        "4. Is the average time users spent on the new page the same for the different language users?"
      ],
      "metadata": {
        "id": "UPzKI4yrQp1k"
      },
      "id": "UPzKI4yrQp1k"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "a6a3d218"
      },
      "source": [
        "##**Import all the necessary libraries**"
      ],
      "id": "a6a3d218"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "658c5dec"
      },
      "source": [
        "# Libraries to help with reading and manipulating data\n",
        "import numpy as np\n",
        "import pandas as pd\n",
        "\n",
        "# Libraries to help with data visualization\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "from matplotlib.patches import Patch\n",
        "from matplotlib.lines import Line2D\n",
        "%matplotlib inline\n",
        "\n",
        "# Library to help with statistical analysis\n",
        "import scipy.stats as stats\n",
        "from statsmodels.stats.proportion import proportions_ztest\n",
        "from scipy.stats import chi2_contingency"
      ],
      "id": "658c5dec",
      "execution_count": 1,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6970dd99"
      },
      "source": [
        "##**Reading the Data into a DataFrame**"
      ],
      "id": "6970dd99"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "d3b95640"
      },
      "source": [
        "# read the dataset\n",
        "data = pd.read_csv('abtest.csv')\n",
        "df = data[:]"
      ],
      "id": "d3b95640",
      "execution_count": 2,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9d75cfa2"
      },
      "source": [
        "##**Explore the dataset and extract insights using Exploratory Data Analysis**"
      ],
      "id": "9d75cfa2"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "27201534"
      },
      "source": [
        "###**Data Overview**\n",
        "The initial steps to get an overview of any dataset is to:\n",
        "\n",
        "* observe the first few rows of the dataset, to check whether the dataset has been loaded properly or not\n",
        "* get information about the number of rows and columns in the dataset\n",
        "* find out the data types of the columns to ensure that data is stored in the preferred format and the value of each property is as expected\n",
        "* check the statistical summary of the dataset to get an overview of the numerical columns of the data"
      ],
      "id": "27201534"
    },
    {
      "cell_type": "markdown",
      "source": [
        "####Observe the first few rows of the dataset, to check whether the dataset has been loaded properly or not"
      ],
      "metadata": {
        "id": "tIcgCjuwYEpN"
      },
      "id": "tIcgCjuwYEpN"
    },
    {
      "cell_type": "markdown",
      "source": [
        " **Displaying the first few rows of the dataset**"
      ],
      "metadata": {
        "id": "jazmnXgrZdVK"
      },
      "id": "jazmnXgrZdVK"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "e1ad11d4",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "outputId": "fdfeda6d-1d39-4059-e7ac-28716a7134f1"
      },
      "source": [
        "# view the first five rows of the dataset\n",
        "df.head()"
      ],
      "id": "e1ad11d4",
      "execution_count": 3,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "   user_id      group landing_page  time_spent_on_the_page converted  \\\n",
              "0   546592    control          old                    3.48        no   \n",
              "1   546468  treatment          new                    7.13       yes   \n",
              "2   546462  treatment          new                    4.40        no   \n",
              "3   546567    control          old                    3.02        no   \n",
              "4   546459  treatment          new                    4.75       yes   \n",
              "\n",
              "  language_preferred  \n",
              "0            Spanish  \n",
              "1            English  \n",
              "2            Spanish  \n",
              "3             French  \n",
              "4            Spanish  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-850e8f66-655f-4d77-a4eb-2c65255319d1\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>user_id</th>\n",
              "      <th>group</th>\n",
              "      <th>landing_page</th>\n",
              "      <th>time_spent_on_the_page</th>\n",
              "      <th>converted</th>\n",
              "      <th>language_preferred</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>546592</td>\n",
              "      <td>control</td>\n",
              "      <td>old</td>\n",
              "      <td>3.48</td>\n",
              "      <td>no</td>\n",
              "      <td>Spanish</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>546468</td>\n",
              "      <td>treatment</td>\n",
              "      <td>new</td>\n",
              "      <td>7.13</td>\n",
              "      <td>yes</td>\n",
              "      <td>English</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>546462</td>\n",
              "      <td>treatment</td>\n",
              "      <td>new</td>\n",
              "      <td>4.40</td>\n",
              "      <td>no</td>\n",
              "      <td>Spanish</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>546567</td>\n",
              "      <td>control</td>\n",
              "      <td>old</td>\n",
              "      <td>3.02</td>\n",
              "      <td>no</td>\n",
              "      <td>French</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>546459</td>\n",
              "      <td>treatment</td>\n",
              "      <td>new</td>\n",
              "      <td>4.75</td>\n",
              "      <td>yes</td>\n",
              "      <td>Spanish</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-850e8f66-655f-4d77-a4eb-2c65255319d1')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-850e8f66-655f-4d77-a4eb-2c65255319d1 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-850e8f66-655f-4d77-a4eb-2c65255319d1');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 3
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Displaying the last few rows of the dataset**"
      ],
      "metadata": {
        "id": "9sx22GDIYHKO"
      },
      "id": "9sx22GDIYHKO"
    },
    {
      "cell_type": "code",
      "source": [
        "# view the last five rows of the dataset\n",
        "df.tail()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 206
        },
        "id": "Q6FUhNYF0ab2",
        "outputId": "acc4935f-881b-4716-adf2-d2af27788d0d"
      },
      "id": "Q6FUhNYF0ab2",
      "execution_count": 4,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "    user_id      group landing_page  time_spent_on_the_page converted  \\\n",
              "95   546446  treatment          new                    5.15        no   \n",
              "96   546544    control          old                    6.52       yes   \n",
              "97   546472  treatment          new                    7.07       yes   \n",
              "98   546481  treatment          new                    6.20       yes   \n",
              "99   546483  treatment          new                    5.86       yes   \n",
              "\n",
              "   language_preferred  \n",
              "95            Spanish  \n",
              "96            English  \n",
              "97            Spanish  \n",
              "98            Spanish  \n",
              "99            English  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-8a42bdab-0aa4-4c3f-b8cb-131b69a996ce\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>user_id</th>\n",
              "      <th>group</th>\n",
              "      <th>landing_page</th>\n",
              "      <th>time_spent_on_the_page</th>\n",
              "      <th>converted</th>\n",
              "      <th>language_preferred</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>95</th>\n",
              "      <td>546446</td>\n",
              "      <td>treatment</td>\n",
              "      <td>new</td>\n",
              "      <td>5.15</td>\n",
              "      <td>no</td>\n",
              "      <td>Spanish</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>96</th>\n",
              "      <td>546544</td>\n",
              "      <td>control</td>\n",
              "      <td>old</td>\n",
              "      <td>6.52</td>\n",
              "      <td>yes</td>\n",
              "      <td>English</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>97</th>\n",
              "      <td>546472</td>\n",
              "      <td>treatment</td>\n",
              "      <td>new</td>\n",
              "      <td>7.07</td>\n",
              "      <td>yes</td>\n",
              "      <td>Spanish</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>98</th>\n",
              "      <td>546481</td>\n",
              "      <td>treatment</td>\n",
              "      <td>new</td>\n",
              "      <td>6.20</td>\n",
              "      <td>yes</td>\n",
              "      <td>Spanish</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>99</th>\n",
              "      <td>546483</td>\n",
              "      <td>treatment</td>\n",
              "      <td>new</td>\n",
              "      <td>5.86</td>\n",
              "      <td>yes</td>\n",
              "      <td>English</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-8a42bdab-0aa4-4c3f-b8cb-131b69a996ce')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-8a42bdab-0aa4-4c3f-b8cb-131b69a996ce button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-8a42bdab-0aa4-4c3f-b8cb-131b69a996ce');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 4
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "####Get information about the number of rows and columns in the dataset"
      ],
      "metadata": {
        "id": "3gYWmCB6Zlpe"
      },
      "id": "3gYWmCB6Zlpe"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Checking the shape of the dataset**"
      ],
      "metadata": {
        "id": "pPiXBoWi1BLm"
      },
      "id": "pPiXBoWi1BLm"
    },
    {
      "cell_type": "code",
      "source": [
        "# view the shape of the dataset\n",
        "df.shape"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "puU8u3ol0ag-",
        "outputId": "1c343cbb-4408-492c-935e-d30e60a88223"
      },
      "id": "puU8u3ol0ag-",
      "execution_count": 5,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(100, 6)"
            ]
          },
          "metadata": {},
          "execution_count": 5
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "The dataset contains 100 rows and 6 columns."
      ],
      "metadata": {
        "id": "cm42KKhv3P2D"
      },
      "id": "cm42KKhv3P2D"
    },
    {
      "cell_type": "markdown",
      "source": [
        "####Find out the data types of the columns to ensure that data is stored in the preferred format and the value of each property is as expected."
      ],
      "metadata": {
        "id": "kiaXXaLTZ34e"
      },
      "id": "kiaXXaLTZ34e"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Checking the data types of the columns for the dataset**"
      ],
      "metadata": {
        "id": "Ztrrbhzz2KTg"
      },
      "id": "Ztrrbhzz2KTg"
    },
    {
      "cell_type": "code",
      "source": [
        "# check the data types of the columns in the dataset\n",
        "df.info()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fcTQXGRh0alF",
        "outputId": "079d46ce-933c-4698-bff5-62fa5bd51428"
      },
      "id": "fcTQXGRh0alF",
      "execution_count": 6,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "<class 'pandas.core.frame.DataFrame'>\n",
            "RangeIndex: 100 entries, 0 to 99\n",
            "Data columns (total 6 columns):\n",
            " #   Column                  Non-Null Count  Dtype  \n",
            "---  ------                  --------------  -----  \n",
            " 0   user_id                 100 non-null    int64  \n",
            " 1   group                   100 non-null    object \n",
            " 2   landing_page            100 non-null    object \n",
            " 3   time_spent_on_the_page  100 non-null    float64\n",
            " 4   converted               100 non-null    object \n",
            " 5   language_preferred      100 non-null    object \n",
            "dtypes: float64(1), int64(1), object(4)\n",
            "memory usage: 4.8+ KB\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "There are 4 categorical columns and two numeric columns in the dataset."
      ],
      "metadata": {
        "id": "UC6cAhW33V5d"
      },
      "id": "UC6cAhW33V5d"
    },
    {
      "cell_type": "markdown",
      "source": [
        "####Check the statistical summary of the dataset to get an overview of the numerical columns of the data"
      ],
      "metadata": {
        "id": "SC23qH2PZ-ZC"
      },
      "id": "SC23qH2PZ-ZC"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Getting the statistical summary for the numerical variables**"
      ],
      "metadata": {
        "id": "JQ6_qTqyY2RA"
      },
      "id": "JQ6_qTqyY2RA"
    },
    {
      "cell_type": "code",
      "source": [
        "# Statistical summary of the numerical variables\n",
        "df.describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 300
        },
        "id": "yUj8XT4X0n0h",
        "outputId": "8054ca93-d6d3-4574-ce83-ce327c2603d0"
      },
      "id": "yUj8XT4X0n0h",
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "             user_id  time_spent_on_the_page\n",
              "count     100.000000              100.000000\n",
              "mean   546517.000000                5.377800\n",
              "std        52.295779                2.378166\n",
              "min    546443.000000                0.190000\n",
              "25%    546467.750000                3.880000\n",
              "50%    546492.500000                5.415000\n",
              "75%    546567.250000                7.022500\n",
              "max    546592.000000               10.710000"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-10898025-d373-4017-8e4a-86b9b03dbdf1\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>user_id</th>\n",
              "      <th>time_spent_on_the_page</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>100.000000</td>\n",
              "      <td>100.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>546517.000000</td>\n",
              "      <td>5.377800</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>52.295779</td>\n",
              "      <td>2.378166</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>546443.000000</td>\n",
              "      <td>0.190000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>546467.750000</td>\n",
              "      <td>3.880000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>546492.500000</td>\n",
              "      <td>5.415000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>546567.250000</td>\n",
              "      <td>7.022500</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>546592.000000</td>\n",
              "      <td>10.710000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-10898025-d373-4017-8e4a-86b9b03dbdf1')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-10898025-d373-4017-8e4a-86b9b03dbdf1 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-10898025-d373-4017-8e4a-86b9b03dbdf1');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Getting the statistical summary for the categorical variables**"
      ],
      "metadata": {
        "id": "G8rAWhxBcBGv"
      },
      "id": "G8rAWhxBcBGv"
    },
    {
      "cell_type": "code",
      "source": [
        "# Statistical summary of the categorical variables\n",
        "def cat_describe(col):\n",
        "  print(f'Statistical summary of the categorical variable `{col}`:')\n",
        "  display(pd.DataFrame({'frequency': df.value_counts(col), \n",
        "                        'percentage': df.value_counts(col, normalize=True)}))\n",
        "  print('-'*80)\n",
        "\n",
        "for col in df.select_dtypes('object'):\n",
        "  cat_describe(col)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 693
        },
        "id": "oM3vQstDcBUr",
        "outputId": "c5d5cd9d-92f6-4cce-f4a3-0dd6ad2d59fa"
      },
      "id": "oM3vQstDcBUr",
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Statistical summary of the categorical variable `group`:\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "           frequency  percentage\n",
              "group                           \n",
              "control           50         0.5\n",
              "treatment         50         0.5"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-99c75102-a663-4589-af7c-a556dfe9dce8\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>frequency</th>\n",
              "      <th>percentage</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>group</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>control</th>\n",
              "      <td>50</td>\n",
              "      <td>0.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>treatment</th>\n",
              "      <td>50</td>\n",
              "      <td>0.5</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-99c75102-a663-4589-af7c-a556dfe9dce8')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-99c75102-a663-4589-af7c-a556dfe9dce8 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-99c75102-a663-4589-af7c-a556dfe9dce8');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "Statistical summary of the categorical variable `landing_page`:\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "              frequency  percentage\n",
              "landing_page                       \n",
              "new                  50         0.5\n",
              "old                  50         0.5"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-1bb9d1a1-156c-4dff-92f2-022bcd6c3427\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>frequency</th>\n",
              "      <th>percentage</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>landing_page</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>new</th>\n",
              "      <td>50</td>\n",
              "      <td>0.5</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>old</th>\n",
              "      <td>50</td>\n",
              "      <td>0.5</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-1bb9d1a1-156c-4dff-92f2-022bcd6c3427')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-1bb9d1a1-156c-4dff-92f2-022bcd6c3427 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-1bb9d1a1-156c-4dff-92f2-022bcd6c3427');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "Statistical summary of the categorical variable `converted`:\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "           frequency  percentage\n",
              "converted                       \n",
              "yes               54        0.54\n",
              "no                46        0.46"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-72c0931e-450b-4f6c-994d-90d7483d3140\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>frequency</th>\n",
              "      <th>percentage</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>converted</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>yes</th>\n",
              "      <td>54</td>\n",
              "      <td>0.54</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>no</th>\n",
              "      <td>46</td>\n",
              "      <td>0.46</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-72c0931e-450b-4f6c-994d-90d7483d3140')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-72c0931e-450b-4f6c-994d-90d7483d3140 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-72c0931e-450b-4f6c-994d-90d7483d3140');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--------------------------------------------------------------------------------\n",
            "Statistical summary of the categorical variable `language_preferred`:\n"
          ]
        },
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "                    frequency  percentage\n",
              "language_preferred                       \n",
              "French                     34        0.34\n",
              "Spanish                    34        0.34\n",
              "English                    32        0.32"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e570c588-6c97-44cf-b3da-cbf30b637895\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>frequency</th>\n",
              "      <th>percentage</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>language_preferred</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>French</th>\n",
              "      <td>34</td>\n",
              "      <td>0.34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Spanish</th>\n",
              "      <td>34</td>\n",
              "      <td>0.34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>English</th>\n",
              "      <td>32</td>\n",
              "      <td>0.32</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e570c588-6c97-44cf-b3da-cbf30b637895')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-e570c588-6c97-44cf-b3da-cbf30b637895 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-e570c588-6c97-44cf-b3da-cbf30b637895');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {}
        },
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "--------------------------------------------------------------------------------\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Check for missing values**"
      ],
      "metadata": {
        "id": "4v64hNL41Oml"
      },
      "id": "4v64hNL41Oml"
    },
    {
      "cell_type": "code",
      "source": [
        "df.isna().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cD_-cKvQ0lAv",
        "outputId": "a343155c-74d5-4a88-dfde-4a355d4b4030"
      },
      "id": "cD_-cKvQ0lAv",
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "user_id                   0\n",
              "group                     0\n",
              "landing_page              0\n",
              "time_spent_on_the_page    0\n",
              "converted                 0\n",
              "language_preferred        0\n",
              "dtype: int64"
            ]
          },
          "metadata": {},
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "There are no missing values in the data."
      ],
      "metadata": {
        "id": "Mx396Wya4Ksn"
      },
      "id": "Mx396Wya4Ksn"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Check for duplicates**"
      ],
      "metadata": {
        "id": "3L_u__p71TQv"
      },
      "id": "3L_u__p71TQv"
    },
    {
      "cell_type": "code",
      "source": [
        "df.duplicated().sum()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S3LDOD6j0n_q",
        "outputId": "a92da835-c131-4588-e0f6-309aa50274dc"
      },
      "id": "S3LDOD6j0n_q",
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "There are no duplicate rows in the data."
      ],
      "metadata": {
        "id": "dETwA3uT4PLI"
      },
      "id": "dETwA3uT4PLI"
    },
    {
      "cell_type": "code",
      "source": [
        "df['user_id'].nunique()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "fnt8O6GF6WRV",
        "outputId": "84bb4a9d-a394-4dfe-cfc1-779f0b888a65"
      },
      "id": "fnt8O6GF6WRV",
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "100"
            ]
          },
          "metadata": {},
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "`user_id` is unique"
      ],
      "metadata": {
        "id": "Rb9xjSc-g4jP"
      },
      "id": "Rb9xjSc-g4jP"
    },
    {
      "cell_type": "code",
      "source": [
        "# Categorical columns\n",
        "cats = list(df.select_dtypes('object'))\n",
        "\n",
        "#Looking for columns that duplicate data\n",
        "for i in range(0, len(cats)-1):\n",
        "  for j in range(i+1, len(cats)):\n",
        "    if (pd.crosstab(df[cats[i]], df[cats[j]])==0).sum().sum()>0:\n",
        "      display(pd.crosstab(df[cats[i]], df[cats[j]]))"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 143
        },
        "id": "rPCZO6k6ggwS",
        "outputId": "c2326290-f661-4ec0-eeea-aee40f3cf49b"
      },
      "id": "rPCZO6k6ggwS",
      "execution_count": 12,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "landing_page  new  old\n",
              "group                 \n",
              "control         0   50\n",
              "treatment      50    0"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-db71854f-59dc-404a-a7b9-3f4eb85c1725\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>landing_page</th>\n",
              "      <th>new</th>\n",
              "      <th>old</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>group</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>control</th>\n",
              "      <td>0</td>\n",
              "      <td>50</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>treatment</th>\n",
              "      <td>50</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-db71854f-59dc-404a-a7b9-3f4eb85c1725')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-db71854f-59dc-404a-a7b9-3f4eb85c1725 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-db71854f-59dc-404a-a7b9-3f4eb85c1725');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Columns `landing_page` and `group` carry the same information and duplicate each other. \n",
        "  * Control group used old page\n",
        "  * Treatment group used new page\n",
        "* Removing `landing_page` column from the working dataset"
      ],
      "metadata": {
        "id": "Kut6Ic_Vg_3a"
      },
      "id": "Kut6Ic_Vg_3a"
    },
    {
      "cell_type": "code",
      "source": [
        "cats.remove('landing_page')\n",
        "df = df.drop('landing_page', axis=1).set_index('user_id', drop=True)"
      ],
      "metadata": {
        "id": "-fkOZqM9jjuk"
      },
      "id": "-fkOZqM9jjuk",
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "####EDA insights:\n",
        "* The original dataset contains 100 rows and 6 columns\n",
        "  * There are 4 categorical columns and two numeric columns in the dataset\n",
        "  * The `user_id` is a column consisting of unique user-ids\n",
        "  * `landing_page` colum duplicated data in `group` column and was removed\n",
        "* There are no missing values in the data\n",
        "* There are no duplicate rows in the data\n",
        "* The mean time spent by the user on the landing page (5.38 mins) is close to the median time (5.42 mins)\n",
        "* The maximum time users spent on the landing page is 10.71 minutes, minimum is 0.19 mins"
      ],
      "metadata": {
        "id": "Kc9Vsq8M3HR-"
      },
      "id": "Kc9Vsq8M3HR-"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "68f3b2c9"
      },
      "source": [
        "###**Univariate Analysis**\n"
      ],
      "id": "68f3b2c9"
    },
    {
      "cell_type": "markdown",
      "source": [
        "Analysing the `time_spent_on_the_page` column\n"
      ],
      "metadata": {
        "id": "YBhVCDrWPm2O"
      },
      "id": "YBhVCDrWPm2O"
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(nrows=2, ncols=1, figsize=(12, 6), sharex=True)\n",
        "fig.suptitle('Time (in minutes) spent by the user on the landing page', fontsize=16); \n",
        "\n",
        "sns.boxplot(data=df, x='time_spent_on_the_page', showmeans=True, ax=ax[0]);\n",
        "sns.histplot(data=df, x='time_spent_on_the_page', ax=ax[1], kde=True);\n",
        "\n",
        "ax[0].set_xlabel('')\n",
        "ax[1].set_xlabel('Time spent, min')\n",
        "ax[0].set_ylabel('')\n",
        "ax[1].set_ylabel('Number of users');\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 431
        },
        "id": "kyAXv0IZ-qqA",
        "outputId": "9cc63d1e-667d-4258-c5fc-7c2404ad5f18"
      },
      "id": "kyAXv0IZ-qqA",
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x432 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Creating list for the columns' names\n",
        "vars = {'time_spent_on_the_page': 'Time spent',\n",
        "        'group': 'Group', \n",
        "        'landing_page': 'Page',\n",
        "        'converted': 'Converted',\n",
        "        'language_preferred': 'Language'}"
      ],
      "metadata": {
        "id": "9b-TU0frYZUF"
      },
      "id": "9b-TU0frYZUF",
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Categorical columns value counts\n",
        "fig, ax = plt.subplots(nrows=1, ncols=len(cats), figsize=(12, 5))\n",
        "fig.suptitle('Categorical columns - value counts', fontsize=18);\n",
        "colors = sns.color_palette('pastel')[0:3]\n",
        "\n",
        "for i, row in enumerate(cats):\n",
        "  ax[i].pie(df.value_counts(row), labels=df.value_counts(row).index, colors=colors, autopct='%.0f%%', textprops={'fontsize': 12})\n",
        "  ax[i].add_artist(plt.Circle((0, 0), 0.25, fc='white'))\n",
        "  ax[i].set_title(vars[row], fontsize=16)\n",
        "  ax[i].set_ylabel('')\n",
        "  ax[i].set_xlabel('')\n",
        "\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 301
        },
        "id": "pKnlSQ1-Ps6p",
        "outputId": "4f0f09ff-2600-4182-d62d-b0ab3ffbbe69"
      },
      "id": "pKnlSQ1-Ps6p",
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x360 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Univariate Analysis conclusions:**\n",
        "* Distribution of the time spent by the user on the landing page is symmetric bell shaped, unimodal, no outliers were observed\n",
        "* There were 2 groups: `control` and `treatment` by 50 observations (50\\%) in each group\n",
        "* 54 users converted to a subscriber of the news portal, and 46 - did not.\n",
        "* There were 3 languages to view the landing page: French, Spanish, and English - 34/34/32 users for each accordingly\n",
        "* There was only one numeric column `time_spent_on_the_page`\n",
        "\n",
        "|index|time\\_spent\\_on\\_the\\_page|\n",
        "|---|---|\n",
        "|count|100\\.0|\n",
        "|mean|5\\.3778|\n",
        "|std|2\\.3782|\n",
        "|min|0\\.19|\n",
        "|25%|3\\.88|\n",
        "|50%|5\\.415|\n",
        "|75%|7\\.0225|\n",
        "|max|10\\.71|\n"
      ],
      "metadata": {
        "id": "BABGUvAEPHAS"
      },
      "id": "BABGUvAEPHAS"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "fad2de2f"
      },
      "source": [
        "###**Bivariate Analysis**"
      ],
      "id": "fad2de2f"
    },
    {
      "cell_type": "code",
      "source": [
        "# time_spent_on_the_page by groups\n",
        "bi = []\n",
        "for c in cats:\n",
        "  for v in df[c].unique():\n",
        "    bi.append([c, v,\n",
        "               f'{df[df[c] == v].time_spent_on_the_page.mean():.2f}',\n",
        "               f'{df[df[c] == v].time_spent_on_the_page.std():.2f}',\n",
        "               f'{stats.iqr(df[df[c] == v].time_spent_on_the_page):.2f}',\n",
        "               f'{df[df[c] == v].time_spent_on_the_page.count():d}'])\n",
        "bi = pd.DataFrame(data=bi, columns=['var', 'value', 'mean', 'std', 'iqr', 'count'])\n",
        "bi"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "id": "IXTgOIMfYN1G",
        "outputId": "132ee298-28e1-494f-c696-a6e159b5f0c6"
      },
      "id": "IXTgOIMfYN1G",
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "                  var      value  mean   std   iqr count\n",
              "0               group    control  4.53  2.58  3.72    50\n",
              "1               group  treatment  6.22  1.82  1.98    50\n",
              "2           converted         no  3.92  2.23  2.59    46\n",
              "3           converted        yes  6.62  1.71  1.87    54\n",
              "4  language_preferred    Spanish  5.33  1.82  2.04    34\n",
              "5  language_preferred    English  5.56  2.62  3.52    32\n",
              "6  language_preferred     French  5.25  2.68  3.97    34"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-83573e03-2253-4c30-87cd-9704af47418c\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>var</th>\n",
              "      <th>value</th>\n",
              "      <th>mean</th>\n",
              "      <th>std</th>\n",
              "      <th>iqr</th>\n",
              "      <th>count</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>group</td>\n",
              "      <td>control</td>\n",
              "      <td>4.53</td>\n",
              "      <td>2.58</td>\n",
              "      <td>3.72</td>\n",
              "      <td>50</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>group</td>\n",
              "      <td>treatment</td>\n",
              "      <td>6.22</td>\n",
              "      <td>1.82</td>\n",
              "      <td>1.98</td>\n",
              "      <td>50</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>converted</td>\n",
              "      <td>no</td>\n",
              "      <td>3.92</td>\n",
              "      <td>2.23</td>\n",
              "      <td>2.59</td>\n",
              "      <td>46</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>converted</td>\n",
              "      <td>yes</td>\n",
              "      <td>6.62</td>\n",
              "      <td>1.71</td>\n",
              "      <td>1.87</td>\n",
              "      <td>54</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>language_preferred</td>\n",
              "      <td>Spanish</td>\n",
              "      <td>5.33</td>\n",
              "      <td>1.82</td>\n",
              "      <td>2.04</td>\n",
              "      <td>34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>language_preferred</td>\n",
              "      <td>English</td>\n",
              "      <td>5.56</td>\n",
              "      <td>2.62</td>\n",
              "      <td>3.52</td>\n",
              "      <td>32</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>language_preferred</td>\n",
              "      <td>French</td>\n",
              "      <td>5.25</td>\n",
              "      <td>2.68</td>\n",
              "      <td>3.97</td>\n",
              "      <td>34</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-83573e03-2253-4c30-87cd-9704af47418c')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-83573e03-2253-4c30-87cd-9704af47418c button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-83573e03-2253-4c30-87cd-9704af47418c');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "58e9a1d0",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 398
        },
        "outputId": "5feb77b5-cd95-4a7b-e93e-6ab4f099baaf"
      },
      "source": [
        "# Distributions of numeric and categorical variables in the dataset\n",
        "fig, ax = plt.subplots(nrows=1, ncols=len(cats), figsize=(12, 5))\n",
        "fig.suptitle('Distributions of numeric and categorical variables in the dataset', fontsize=18, y=1.05);\n",
        "\n",
        "for i, row in enumerate(cats):\n",
        "  sns.boxplot(y=df['time_spent_on_the_page'], x=df[row], ax=ax[i], orient='v')\n",
        "  ax[i].set_title('Time spent vs ' + vars[row], fontsize=12)\n",
        "  ax[i].set_ylabel('Time spent (min)', fontsize=12)\n",
        "  ax[i].set_xlabel(vars[row], fontsize=12)\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show();"
      ],
      "id": "58e9a1d0",
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x360 with 3 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Time spent by language and group\n",
        "fig, ax = plt.subplots(1, 2, figsize=(16, 6))\n",
        "fig.suptitle('Time spent by language and group', fontsize=18);\n",
        "\n",
        "sns.boxplot(y=df['time_spent_on_the_page'], x=df['group'], hue=df['language_preferred'], ax=ax[0], orient='v')\n",
        "sns.boxplot(y=df['time_spent_on_the_page'], x=df['language_preferred'], hue=df['group'], ax=ax[1], orient='v')\n",
        "ax[0].set_ylabel('Time spent (min)')\n",
        "ax[0].set_xlabel('Group')\n",
        "ax[1].set_ylabel('Time spent (min)')\n",
        "ax[1].set_xlabel('Language')\n",
        "\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 414
        },
        "id": "fMJ5zu7neA3p",
        "outputId": "bc8d95fd-3472-45d5-c896-c0f2c67a682d"
      },
      "id": "fMJ5zu7neA3p",
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1152x432 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Time spent by language and group\n",
        "fig, ax = plt.subplots(1, 2, figsize=(16, 6))\n",
        "\n",
        "sns.boxplot(y=df['time_spent_on_the_page'], x=df['group'], hue=df['converted'], ax=ax[0], orient='v')\n",
        "sns.boxplot(y=df['time_spent_on_the_page'], x=df['language_preferred'], hue=df['converted'], ax=ax[1], orient='v')\n",
        "ax[0].set_title('Time spent by group and conversion', fontsize=14);\n",
        "ax[0].set_ylabel('Time spent (min)')\n",
        "ax[0].set_xlabel('Group')\n",
        "ax[1].set_title('Time spent by language and conversion', fontsize=14);\n",
        "ax[1].set_ylabel('Time spent (min)')\n",
        "ax[1].set_xlabel('Language')\n",
        "\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 389
        },
        "id": "qiW9uF-xgJ0u",
        "outputId": "350a454c-bbbe-4a04-bcb1-b1d7e542ee2f"
      },
      "id": "qiW9uF-xgJ0u",
      "execution_count": 20,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1152x432 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_conv = df.replace({'converted': {'no': 0, 'yes': 1}}).groupby(['group', 'language_preferred']).mean()[['converted']].reset_index()\n",
        "\n",
        "fig, ax = plt.subplots(1, 2, figsize=(16, 6))\n",
        "fig.suptitle('Conversion rate by language and group', fontsize=18);\n",
        "\n",
        "sns.barplot(data=df_conv,y='converted', x='language_preferred', hue='group', ax=ax[0])\n",
        "ax[0].set_ylabel('Coversion rate')\n",
        "ax[0].set_xlabel('Group')\n",
        "\n",
        "sns.barplot(data=df_conv, y='converted', x='group', hue='language_preferred', ax=ax[1])\n",
        "ax[1].set_ylabel('Coversion rate')\n",
        "ax[1].set_xlabel('Language')\n",
        "ax[1].legend(loc='upper center')\n",
        "\n",
        "plt.show();"
      ],
      "metadata": {
        "id": "zSl7OUcZlbtr",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 413
        },
        "outputId": "0c9dc056-5cb7-49a4-91ef-8e278b384d95"
      },
      "id": "zSl7OUcZlbtr",
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 1152x432 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "df_conv.pivot_table(index='group', columns='language_preferred', values='converted', margins=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "YZXXGXcFNtdP",
        "outputId": "cf98aba2-5761-492f-9614-17b32bd26c9a"
      },
      "id": "YZXXGXcFNtdP",
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "language_preferred  English    French   Spanish       All\n",
              "group                                                    \n",
              "control             0.68750  0.176471  0.411765  0.425245\n",
              "treatment           0.62500  0.705882  0.647059  0.659314\n",
              "All                 0.65625  0.441176  0.529412  0.542279"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-ad5bc260-b01d-4f09-b374-35503b31b237\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>language_preferred</th>\n",
              "      <th>English</th>\n",
              "      <th>French</th>\n",
              "      <th>Spanish</th>\n",
              "      <th>All</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>group</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>control</th>\n",
              "      <td>0.68750</td>\n",
              "      <td>0.176471</td>\n",
              "      <td>0.411765</td>\n",
              "      <td>0.425245</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>treatment</th>\n",
              "      <td>0.62500</td>\n",
              "      <td>0.705882</td>\n",
              "      <td>0.647059</td>\n",
              "      <td>0.659314</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>All</th>\n",
              "      <td>0.65625</td>\n",
              "      <td>0.441176</td>\n",
              "      <td>0.529412</td>\n",
              "      <td>0.542279</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-ad5bc260-b01d-4f09-b374-35503b31b237')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-ad5bc260-b01d-4f09-b374-35503b31b237 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-ad5bc260-b01d-4f09-b374-35503b31b237');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Bivariate Analysis conclusions:**\n",
        "* **Mean time**\n",
        "  * The mean time in the treatment group (6.22') is greater than in the control group (4.53')\n",
        "    * Standard deviation & IQR is less for the treatment group\n",
        "    * There are a few outliers in the treatment group only\n",
        "  * Mean time for converted users (6.62') is greater than not converted (3.92')\n",
        "    * Standard deviation & IQR is less for those who converted\n",
        "    * There are a few outliers in the 'converted' group only\n",
        "    * IQRs are not overlapped (Q1-converted is greater than Q3-not-converted)\n",
        "  * Mean times for different languages are approximately the same (5.25' - 5.56')\n",
        "    * Standard deviation and IQR for Spanish are less than for English and French\n",
        "    * There is an outlier in the Spanish group only\n",
        "* **Conversion rate**\n",
        "  * Conversion rate of the Spanish and especially French-language users increased significantly on the new landing page (~1.5 times and ~3.5 times respectively)\n",
        "  * Conversion rate of the English-language users did not change significantly on the new landing page\n",
        "  * Conversion rate on the existing page varied from ~0.2 to 0.7, depending on the language\n",
        "  * Conversion rate on the new page is approximately 0.65 for all languages\n",
        "  * Conversion rate for the French language has improved significantly on the new landing page\n",
        "* The average time converted users spent on the landing page is higher for all the groups and languages: Q1 of time converted users is consistently higher than Q3 of time non-converted users\n"
      ],
      "metadata": {
        "id": "d40DYVVnUe7C"
      },
      "id": "d40DYVVnUe7C"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "27906576"
      },
      "source": [
        "##**1. Do the users spend more time on the new landing page than the existing landing page?**"
      ],
      "id": "27906576"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "e5fdf0c8"
      },
      "source": [
        "### Perform Visual Analysis"
      ],
      "id": "e5fdf0c8"
    },
    {
      "cell_type": "code",
      "source": [
        "df.groupby('group').describe()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "Y80GfhVryhhb",
        "outputId": "81c32d1b-e20f-4367-d1f6-38ed1dad951b"
      },
      "id": "Y80GfhVryhhb",
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "          time_spent_on_the_page                                        \\\n",
              "                           count    mean       std   min    25%    50%   \n",
              "group                                                                    \n",
              "control                     50.0  4.5324  2.581975  0.19  2.720  4.380   \n",
              "treatment                   50.0  6.2232  1.817031  1.65  5.175  6.105   \n",
              "\n",
              "                          \n",
              "              75%    max  \n",
              "group                     \n",
              "control    6.4425  10.30  \n",
              "treatment  7.1600  10.71  "
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-e918e7b8-6b0b-42a9-a163-aad17e824aba\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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\">time_spent_on_the_page</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>group</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>control</th>\n",
              "      <td>50.0</td>\n",
              "      <td>4.5324</td>\n",
              "      <td>2.581975</td>\n",
              "      <td>0.19</td>\n",
              "      <td>2.720</td>\n",
              "      <td>4.380</td>\n",
              "      <td>6.4425</td>\n",
              "      <td>10.30</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>treatment</th>\n",
              "      <td>50.0</td>\n",
              "      <td>6.2232</td>\n",
              "      <td>1.817031</td>\n",
              "      <td>1.65</td>\n",
              "      <td>5.175</td>\n",
              "      <td>6.105</td>\n",
              "      <td>7.1600</td>\n",
              "      <td>10.71</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-e918e7b8-6b0b-42a9-a163-aad17e824aba')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-e918e7b8-6b0b-42a9-a163-aad17e824aba button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-e918e7b8-6b0b-42a9-a163-aad17e824aba');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6eZJa41eg21n",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 441
        },
        "outputId": "8d83b4b1-59a9-44cf-bbba-e085b8799c17"
      },
      "source": [
        "# Boxplot of `time_spent_on_the_page` by groups\n",
        "fig, ax = plt.subplots(figsize=(6, 6)) \n",
        "sns.boxplot(y=df['time_spent_on_the_page'], x=df['group'])\n",
        "ax.set_title('Time spent by groups', fontsize=14)\n",
        "ax.set_ylabel('Time spent (min)', fontsize=14)\n",
        "ax.set_xlabel('Group', fontsize=14)\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show();"
      ],
      "id": "6eZJa41eg21n",
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* According to Visual Analysis, the users spend more time on the new landing page (treatment group) than on the existing landing page (control group)\n",
        "* A higher standard deviation of the time spent in the existing landing page (control group) indicates data are more spread out\n",
        "  *Contrary, data points of the treatment group are close to the mean\n",
        "* The median time of the treatment group is within interquartile interval of the control group\n",
        "* The median time of the control group is below Q1 of the treatment group"
      ],
      "metadata": {
        "id": "88L9j9DMqEFn"
      },
      "id": "88L9j9DMqEFn"
    },
    {
      "cell_type": "markdown",
      "source": [
        "###**Hypothesis Testing**"
      ],
      "metadata": {
        "id": "jCkkCphQzu2H"
      },
      "id": "jCkkCphQzu2H"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "56b3dafd"
      },
      "source": [
        "### Step 1: Define the null and alternate hypotheses"
      ],
      "id": "56b3dafd"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "d2ac77ef"
      },
      "source": [
        "**Let's write the null and alternative hypotheses**\n",
        "\n",
        ">$H_0$: The mean time users spent on the new landing page is equal to the time users spent on the existing page. <br>\n",
        ">$H_a$: The mean time users spent on the new landing page is greater than that from the existing landing page.\n",
        "\n",
        "Let $\\mu_1, \\mu_2$ be the mean time from the new landing page and the mean time from the existing landing page, respectively. We will test the null hypothesis.\n",
        "\n",
        "Mathematically, the above-formulated hypotheses can be written as follows:\n",
        "\n",
        ">$H_0:\\mu_1=\\mu_2$ <br>\n",
        ">$H_a:\\mu_1>\\mu_2$"
      ],
      "id": "d2ac77ef"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "c7ee4907"
      },
      "source": [
        "### Step 2: Select Appropriate test"
      ],
      "id": "c7ee4907"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "2c183cd7"
      },
      "source": [
        "The Data Science team experimented by randomly selecting 100 users and dividing them equally into two groups. The existing landing page was served to the first group (control group) and the new landing page to the second group (treatment group). Data regarding the time users in both groups spent with the two versions of the landing page was collected. We need to confirm that the users spend more time on the new landing page than the existing landing page.\n",
        "\n",
        "This is a case of a one-tailed test comparing means from two populations. As we are comparing two means calculated from different sets of users, the two groups are independent. Also, the populations' standard deviations are unknown.\n",
        "\n",
        ">Hence, a **Two Independent Sample T-test for Equality of Means (unequal standard deviations)** is the most appropriate test for this case."
      ],
      "id": "2c183cd7"
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(figsize=(8, 6))\n",
        "fig.suptitle('Time (in minutes) users spent on the landing page', fontsize=16); \n",
        "sns.histplot(data=df, x='time_spent_on_the_page', hue='group', ax=ax, kde=True);\n",
        "ax.set_xlabel('Time spent, min', fontsize=14)\n",
        "ax.set_ylabel('Number of users', fontsize=14);\n",
        "plt.show()"
      ],
      "metadata": {
        "id": "DLipJyTDmMPr",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 435
        },
        "outputId": "e7a5d18b-6103-460e-8ece-6bf1835d6401"
      },
      "id": "DLipJyTDmMPr",
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fig, ax = plt.subplots(nrows=2, ncols=2, figsize=(12, 6))\n",
        "fig.suptitle('Time (in minutes) spent by the user on the landing page', fontsize=16); \n",
        "\n",
        "dfc = df[df['group']=='control']\n",
        "dft = df[df['group']=='treatment']\n",
        "\n",
        "sns.boxplot(data=dfc, x='time_spent_on_the_page', ax=ax[0,0]);\n",
        "sns.histplot(data=dfc, x='time_spent_on_the_page', ax=ax[1,0], kde=True);\n",
        "\n",
        "sns.boxplot(data=dft, x='time_spent_on_the_page', ax=ax[0,1]);\n",
        "sns.histplot(data=dft, x='time_spent_on_the_page', ax=ax[1,1], kde=True);\n",
        "\n",
        "ax[0,0].set_title('Existing landing page (control group)')\n",
        "ax[0,0].set_xlabel('')\n",
        "ax[1,0].set_xlabel('Time spent, min')\n",
        "ax[1,0].set_ylabel('Number of users');\n",
        "\n",
        "ax[0,1].set_title('New landing page (treatment group)')\n",
        "ax[0,1].set_xlabel('')\n",
        "ax[1,1].set_xlabel('Time spent, min')\n",
        "ax[1,1].set_ylabel('Number of users');\n",
        "\n",
        "plt.show()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 431
        },
        "id": "ONZ8CatzUxS2",
        "outputId": "52a702c1-733f-416e-de94-a7842e42bb0f"
      },
      "id": "ONZ8CatzUxS2",
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x432 with 4 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Sample statistics ( $\\mu$ and $\\sigma$ ) calculations**"
      ],
      "metadata": {
        "id": "R-xBU3Gddtl8"
      },
      "id": "R-xBU3Gddtl8"
    },
    {
      "cell_type": "code",
      "source": [
        "print(f'Mean of time users spent on the old page is {dfc.time_spent_on_the_page.mean():.4f}, standard deviation of the sample is {dfc.time_spent_on_the_page.std():.4f}')\n",
        "print(f'Mean of time users spent on the new page is {dft.time_spent_on_the_page.mean():.4f}, standard deviation of the sample is {dft.time_spent_on_the_page.std():.4f}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "QsdZyV3odVmp",
        "outputId": "5cf10b2c-7fe8-4cbc-9947-6537950e23a1"
      },
      "id": "QsdZyV3odVmp",
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Mean of time users spent on the old page is 4.5324, standard deviation of the sample is 2.5820\n",
            "Mean of time users spent on the new page is 6.2232, standard deviation of the sample is 1.8170\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Shapiro-Wilk’s test** (Normally distributed populations)\n",
        "\n",
        "We will test the null hypothesis\n",
        "\n",
        ">$H_0:$ The time users spent on the landing page follow a normal distribution\n",
        "\n",
        "against the alternative hypothesis\n",
        "\n",
        ">$H_a:$ The time users spent on the landing page do not not follow a normal distribution"
      ],
      "metadata": {
        "id": "oJebhL9wYeP0"
      },
      "id": "oJebhL9wYeP0"
    },
    {
      "cell_type": "code",
      "source": [
        "pvc = stats.shapiro(dfc.time_spent_on_the_page).pvalue\n",
        "pvt = stats.shapiro(dft.time_spent_on_the_page).pvalue\n",
        "alpha = 0.05\n",
        "\n",
        "print(f'The p-value for the existing page is {pvc:.4f}')\n",
        "print(f'The p-value for the new page is {pvt:.4f}\\n')\n",
        "\n",
        "print('Normality of the control group:')\n",
        "if (pvc < alpha): print(f'As the p-value {pvc:.2f} is less than the level of significance {alpha}, we reject the null hypothesis. ')\n",
        "else: print(f'As the p-value {pvc:.2f} is greater than the level of significance {alpha}, we fail to reject the null hypothesis.')\n",
        "\n",
        "print('Normality of the tretment group:')\n",
        "if (pvt < alpha): print(f'As the p-value {pvt:.2f} is less than the level of significance {alpha}, we reject the null hypothesis. ')\n",
        "else: print(f'As the p-value {pvt:.2f} is greater than the level of significance {alpha}, we fail to reject the null hypothesis.')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KoKWaxj5XcoK",
        "outputId": "8dd6d51d-7e92-488d-c785-edf171290fda"
      },
      "id": "KoKWaxj5XcoK",
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value for the existing page is 0.4561\n",
            "The p-value for the new page is 0.8040\n",
            "\n",
            "Normality of the control group:\n",
            "As the p-value 0.46 is greater than the level of significance 0.05, we fail to reject the null hypothesis.\n",
            "Normality of the tretment group:\n",
            "As the p-value 0.80 is greater than the level of significance 0.05, we fail to reject the null hypothesis.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        ">Since both p-values of the tests are greater than the 5% significance level,\n",
        "we fail to reject the null hypothesis that the time users spent on\n",
        "the landing page follows the normal distribution for both control and treatment groups. **The distributions follow normal distribution**"
      ],
      "metadata": {
        "id": "WQJBGGdLCHOE"
      },
      "id": "WQJBGGdLCHOE"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Levene’s test** (Homogeneity of variances)\n",
        "\n",
        "We will test the null hypothesis\n",
        "\n",
        ">$H_0$: All the population variances are equal\n",
        "\n",
        "against the alternative hypothesis\n",
        "\n",
        ">$H_a$: At least one variance is different from the rest"
      ],
      "metadata": {
        "id": "5HKSJSOwfXVV"
      },
      "id": "5HKSJSOwfXVV"
    },
    {
      "cell_type": "code",
      "source": [
        "pv = stats.levene(dfc.time_spent_on_the_page, dft.time_spent_on_the_page).pvalue\n",
        "alpha = 0.05\n",
        "\n",
        "print(f'The p-value is {pv:.4f}\\n')\n",
        "\n",
        "if (pv < alpha): print(f'As the p-value {pv:.2f} is less than the level of significance {alpha}, we reject the null hypothesis. ')\n",
        "else: print(f'As the p-value {pv:.2f} is greater than the level of significance {alpha}, we fail to reject the null hypothesis.')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "LtiUZk2CfXqN",
        "outputId": "0e63094c-4d2d-477f-84d2-1707630d28aa"
      },
      "id": "LtiUZk2CfXqN",
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.0090\n",
            "\n",
            "As the p-value 0.01 is less than the level of significance 0.05, we reject the null hypothesis. \n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        ">Since p-value of the test is less than the 5% significance level, we can reject the null hypothesis that the variance of time users spent on the new page is equal to the variance of time on the existing page. **The sample standard deviations are different**."
      ],
      "metadata": {
        "id": "ujWzejrvgXXH"
      },
      "id": "ujWzejrvgXXH"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Let's test whether the T-test assumptions are satisfied or not**\n",
        "\n",
        "- [x] `Continuous data` - **Yes**, the time users spent on the landing page is measured on a continuous scale\n",
        "- [x] `Normally distributed populations` - **Yes**, we performed Shapiro-Wilk’s test that allows us to fail to reject the $H_0$, so the populations are assumed to be normal\n",
        "- [x] `Independent populations` - **Yes**, as we are taking random samples for two different group of users, the two samples are from two independent populations\n",
        "- [x] `Unequal population standard deviations` - **Yes**, as the sample standard deviations are different, the population standard deviations may be assumed to be different.\n",
        "- [x] `Random sampling from the population` - **Yes**, we are informed that the collected sample is a simple random sample.\n",
        "\n",
        ">We can use **Two Independent Sample T-test for Equality of Means (unequal standard deviations)** for this problem."
      ],
      "metadata": {
        "id": "MUcaCu8jUo90"
      },
      "id": "MUcaCu8jUo90"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3f58c9a7"
      },
      "source": [
        "### Step 3: Decide the significance level"
      ],
      "id": "3f58c9a7"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "92d7d7ee"
      },
      "source": [
        "As given in the problem statement, we select $\\alpha$ as $0.05$.\n",
        "\n",
        "Test at 0.05 level of significance whether the data provide sufficient evidence to conclude that the users spend more time on the new landing page than on the existing landing page."
      ],
      "id": "92d7d7ee"
    },
    {
      "cell_type": "code",
      "source": [
        "alpha = 0.05"
      ],
      "metadata": {
        "id": "mFrH0sPrpzCD"
      },
      "id": "mFrH0sPrpzCD",
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5fed2eac"
      },
      "source": [
        "### Step 4: Collect and prepare data"
      ],
      "id": "5fed2eac"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "f5a59495"
      },
      "source": [
        "old = dfc.time_spent_on_the_page # existing page (control group)\n",
        "new = dft.time_spent_on_the_page # new page (treatment group)"
      ],
      "id": "f5a59495",
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "f5c5258d"
      },
      "source": [
        "### Step 5: Calculate the p-value"
      ],
      "id": "f5c5258d"
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can use function [`ttest_ind()`](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.ttest_ind.html) of Scipy to compute the test statistic and p-value for this task as this is a test to compare two sample means from two independent populations when the population standard deviations are unknown.\n",
        "Parameters are:\n",
        "- `old`, `new`: two independent samples of scores (time users spent on the existing (old) and new landing pages, respectively);\n",
        "- `eqaul_var`: is `False` as population variance as not equal;\n",
        "- `alternative`: is `less` because the alternative hypoteses is that the `old` is less than `new`"
      ],
      "metadata": {
        "id": "gDJU8gWwnlwO"
      },
      "id": "gDJU8gWwnlwO"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "fac8594a",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0e4610de-1155-4d01-a0ff-3acb622acab9"
      },
      "source": [
        "# find the p-value\n",
        "test_stat, p_value = stats.ttest_ind(old, new, equal_var = False, alternative = 'less')\n",
        "print('The p-value is ', p_value)"
      ],
      "id": "fac8594a",
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is  0.0001392381225166549\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "359b12f8"
      },
      "source": [
        "### Step 6: Compare the p-value with $\\alpha$"
      ],
      "id": "359b12f8"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7be47289",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "32b1d1e3-84de-460d-9114-8940cf02a5d4"
      },
      "source": [
        "if p_value < alpha:\n",
        "  print(f'As the p-value (~{p_value:.5f}) is less than the level of significance, we reject the null hypothesis.')\n",
        "else:\n",
        "  print(f'As the p-value (~{p_value:.5f}) is greater than the level of significance, we fail reject the null hypothesis.')"
      ],
      "id": "7be47289",
      "execution_count": 33,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "As the p-value (~0.00014) is less than the level of significance, we reject the null hypothesis.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "As the p-value (~0.00014) is less than the level of significance (0.05), we reject the null hypothesis."
      ],
      "metadata": {
        "id": "a-dpKkvUpuOB"
      },
      "id": "a-dpKkvUpuOB"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "d9affc84"
      },
      "source": [
        "### Step 7:  Draw inference"
      ],
      "id": "d9affc84"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "6b16dd4b"
      },
      "source": [
        ">At 5% significance level, we reject the null hypothesis.\n",
        ">Hence, we do have enough statistical evidence to say that the mean time users spend on the new landing page is greater that the mean time the users spent on the existing page.\n",
        "\n",
        "**The users spend more time on the new landing page than on the existing landing page.**"
      ],
      "id": "6b16dd4b"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "353e9d24"
      },
      "source": [
        "##**2. Is the conversion rate (the proportion of users who visit the landing page and get converted) for the new page greater than the conversion rate for the old page?**"
      ],
      "id": "353e9d24"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Perform Visual Analysis"
      ],
      "metadata": {
        "id": "K1b135FwR6LK"
      },
      "id": "K1b135FwR6LK"
    },
    {
      "cell_type": "code",
      "source": [
        "# create a function to draw fractions on the charts nicely\n",
        "def make_autopct(values):\n",
        "    def my_autopct(pct):\n",
        "        total = sum(values)\n",
        "        val = int(round(pct*total/100.0))\n",
        "        return f'$\\\\frac{{{val:2.0f}}}{{{total:2.0f}}}$' + f' ({pct:.0f}%)'\n",
        "    return my_autopct"
      ],
      "metadata": {
        "id": "I13UHJnrnaNI"
      },
      "id": "I13UHJnrnaNI",
      "execution_count": 34,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "# Pie charts of the proportions\n",
        "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(8, 4))\n",
        "fig.suptitle('The proportion of users who visit the landing page and get converted', fontsize=14);\n",
        "colors = sns.color_palette('pastel')[0:2]\n",
        "\n",
        "pt = df.pivot_table(index='converted', columns='group', values='time_spent_on_the_page', aggfunc='count')\n",
        "labels = ['not converted', 'converted']\n",
        "\n",
        "ax[0].pie(pt['control'], colors=colors, labels=labels, autopct=make_autopct(pt['control']), textprops={'fontsize': 12})\n",
        "ax[0].add_artist(plt.Circle((0, 0), 0.3, fc='white'))\n",
        "ax[0].set_ylabel('')\n",
        "ax[0].set_xlabel('Existing page', fontsize=16)\n",
        "ax[1].pie(pt['treatment'], colors=colors, labels=labels, autopct=make_autopct(pt['treatment']), textprops={'fontsize': 12})\n",
        "ax[1].add_artist(plt.Circle((0, 0), 0.3, fc='white'))\n",
        "ax[1].set_ylabel('')\n",
        "ax[1].set_xlabel('New page', fontsize=16)\n",
        "plt.show();\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 289
        },
        "id": "QRODsiqiVugN",
        "outputId": "de9e5e31-f1f9-4357-c4b7-fca174d1c5de"
      },
      "id": "QRODsiqiVugN",
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 576x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        ">$p_1=\\frac{33}{50}\\\\\n",
        "p_2=\\frac{21}{50}$\n"
      ],
      "metadata": {
        "id": "ji6yo6nWh9dC"
      },
      "id": "ji6yo6nWh9dC"
    },
    {
      "cell_type": "markdown",
      "source": [
        "Visually and mathematically, the proportion of users from the samples who visited the new landing page and got converted is greater than that on the existing page.\n",
        "\n",
        "$\\frac{33}{50}>\\frac{21}{50}$"
      ],
      "metadata": {
        "id": "u5hDZej2Q8lC"
      },
      "id": "u5hDZej2Q8lC"
    },
    {
      "cell_type": "markdown",
      "source": [
        "###**Hypothesis Testing**"
      ],
      "metadata": {
        "id": "h3lNiwIXCOzB"
      },
      "id": "h3lNiwIXCOzB"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 1: Define the null and alternate hypotheses\n"
      ],
      "metadata": {
        "id": "0OfSUVa6SADN"
      },
      "id": "0OfSUVa6SADN"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Let's write the null and alternative hypothesis**\n",
        "\n",
        "* `group` and `converted` are two categorical variables.\n",
        "* We want to see if the proportion of users who visited the new landing page and get converted (**conversion rate**) is greater than the proportion of users who visited the existing landing page and get converted.\n",
        "\n",
        ">$H_0:$ The conversion rate for new landing page users is equal to the conversion rate for the existing landing page users. <br>\n",
        ">$H_a:$ The conversion rate for new landing page users is significantly greater than the conversion rate for the existing page users.\n",
        "\n",
        "Let $p_1$ and $p_2$ be the proportions of users who visited the new and existing landing pages respectively and get converted. We will test the null hypothesis.\n",
        "\n",
        "Mathematically, the above-formulated hypotheses can be written as follows:\n",
        "\n",
        ">$H_0:p_1=p_2$ <br>\n",
        ">$H_a:p_1>p_2$\n"
      ],
      "metadata": {
        "id": "MWLBf71EYPit"
      },
      "id": "MWLBf71EYPit"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 2: Select Appropriate test"
      ],
      "metadata": {
        "id": "H60I14XhSFp4"
      },
      "id": "H60I14XhSFp4"
    },
    {
      "cell_type": "markdown",
      "source": [
        "The formulated hypotheses are concerned with proportions. A test of proportions can be used to analyse the hypotheses and draw a conclusion. We shall use a **Proportions Z test** for this problem.\n",
        "\n",
        "**Let's test whether the Z-test assumptions are satisfied or not**\n",
        "\n",
        "- [x] Binomially distributed populations - **Yes**, user either gets converted to a subscriber of the news portal or not \n",
        "- [x] Independent populations - **Yes**, as we are taking random samples for two different group of users, the two samples are from two independent populations\n",
        "- [x] Random sampling from the populations - **Yes**, we are informed that the collected sample is a simple random sample\n",
        "- [x] Can the binomial distribution approximated to normal distribution - **Yes**, for binary data, CLT works slower than usual. The standard thing is to check whether np and n(1-p) are greater than or equal to 10. Here, n and p refer to the sample size and sample proportion respectively. \n",
        "\n",
        ">$n_1p_1 = 50 \\cdot \\frac{33}{50} = 33 \\geq 10\\\\\n",
        "n_1(1-p_1) = 50 \\cdot \\frac{50-33}{50} = 17 \\geq 10 \\\\\n",
        "n_2p_2 = 50 \\cdot \\frac{21}{50} = 21 \\geq 10\\\\\n",
        "n_2(1-p_2) = 50 \\cdot \\frac{50-21}{50} = 29 \\geq 10 $\n"
      ],
      "metadata": {
        "id": "5ABddHKoYLHi"
      },
      "id": "5ABddHKoYLHi"
    },
    {
      "cell_type": "markdown",
      "source": [
        ">We can use one-sided 2-sample **Z-test for Proportions** for this problem."
      ],
      "metadata": {
        "id": "IyOW7BsQXJGQ"
      },
      "id": "IyOW7BsQXJGQ"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 3: Decide the significance level"
      ],
      "metadata": {
        "id": "BaZ1iTO4SM8r"
      },
      "id": "BaZ1iTO4SM8r"
    },
    {
      "cell_type": "markdown",
      "source": [
        "As given in the problem statement, we select $\\alpha$ as $0.05$."
      ],
      "metadata": {
        "id": "JeieUKBXfqZ7"
      },
      "id": "JeieUKBXfqZ7"
    },
    {
      "cell_type": "code",
      "source": [
        "alpha = 0.05"
      ],
      "metadata": {
        "id": "9ja-mydemia4"
      },
      "id": "9ja-mydemia4",
      "execution_count": 36,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 4: Collect and prepare data"
      ],
      "metadata": {
        "id": "9xV3aiUeSIcB"
      },
      "id": "9xV3aiUeSIcB"
    },
    {
      "cell_type": "code",
      "source": [
        "# set the counts of converted users\n",
        "converted_users = np.array([pt.loc['yes'].treatment, pt.loc['yes'].control])\n",
        "\n",
        "# set the sample sizes\n",
        "nobs = np.array(pt.sum(axis=0).values)\n",
        "\n",
        "print(f'Counts of converted users: {converted_users}')\n",
        "print(f'Sample sizes: {nobs}')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GWIStMbYkvOx",
        "outputId": "a17c4e5c-61cd-49c2-8d39-cc935d23bbc1"
      },
      "id": "GWIStMbYkvOx",
      "execution_count": 37,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Counts of converted users: [33 21]\n",
            "Sample sizes: [50 50]\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 5: Calculate the p-value"
      ],
      "metadata": {
        "id": "wemBN_roSfm2"
      },
      "id": "wemBN_roSfm2"
    },
    {
      "cell_type": "markdown",
      "source": [
        "We can use function [`proportions_ztest()`](https://www.statsmodels.org/dev/generated/statsmodels.stats.proportion.proportions_ztest.html) of `statsmodels` to compute the test statistic and p-value for this task as this function uses a simple normal test for proportions.\n",
        "\n",
        "Parameters are:\n",
        "- `count`:  is a list `converted_users`, the number of `converted` in nobs observations\n",
        "- `nobs`: is `nobs`, the number of observations\n",
        "- `alternative`: is `larger` because the alternative hypothesis is that proportion of converted users on the new page $p_1$ is greater than proportion of converted users on the existing page $p_2$"
      ],
      "metadata": {
        "id": "TNVRFX_NTtOJ"
      },
      "id": "TNVRFX_NTtOJ"
    },
    {
      "cell_type": "code",
      "source": [
        "# find the p-value\n",
        "test_stat, p_value = proportions_ztest(count=converted_users, nobs=nobs, alternative='larger')\n",
        "print(f'The p-value is {p_value:.5f}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "tItcyGIYkysM",
        "outputId": "5a6211ae-1a6c-4a56-a2cd-8a540a3a666a"
      },
      "id": "tItcyGIYkysM",
      "execution_count": 38,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.00803\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 6: Compare the p-value with $\\alpha$"
      ],
      "metadata": {
        "id": "UHOPXz2PSgDk"
      },
      "id": "UHOPXz2PSgDk"
    },
    {
      "cell_type": "code",
      "source": [
        "if p_value > alpha:\n",
        "  print(f'As the p-value ({p_value:.3f}) is greater than the significance level ({alpha}), we fail to reject the null hypothesis')\n",
        "else:\n",
        "  print(f'As the p-value ({p_value:.3f}) is less than the significance level ({alpha}), we can reject the null hypothesis')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XQuu_MrhmdPd",
        "outputId": "6dddc8da-5865-4ef6-fc2e-7516b5c856d3"
      },
      "id": "XQuu_MrhmdPd",
      "execution_count": 39,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "As the p-value (0.008) is less than the significance level (0.05), we can reject the null hypothesis\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 7:  Draw inference"
      ],
      "metadata": {
        "id": "dITg4OEqSgLV"
      },
      "id": "dITg4OEqSgLV"
    },
    {
      "cell_type": "markdown",
      "source": [
        ">As the p-value is less than the significance level 0.05, we reject the null hypothesis.\n",
        "Hence, we have enough statistical evidence to say that the conversion rate (the proportion of users who visit the landing page and get converted) for the new landing page users is greater than the conversion rate for the existing page users.\n",
        "\n",
        "**The conversion rate for the new page is greater than the conversion rate for the old page.**"
      ],
      "metadata": {
        "id": "NRjxyQPhmZyI"
      },
      "id": "NRjxyQPhmZyI"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "d49bfa2d"
      },
      "source": [
        "##**3. Is the conversion and preferred language are independent or related?**"
      ],
      "id": "d49bfa2d"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Perform Visual Analysis"
      ],
      "metadata": {
        "id": "768BGR4vSsWv"
      },
      "id": "768BGR4vSsWv"
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Since both the concerned variables are categorical, we can use a contingency table and a stacked bar graph to inspect the data visually.\n",
        "* It was not clear from the objectives, but we can assume that the question is about the experience of the treatment group on the new landing page. However, let us perform this analysis for both groups"
      ],
      "metadata": {
        "id": "Q5JcclsjdTLP"
      },
      "id": "Q5JcclsjdTLP"
    },
    {
      "cell_type": "code",
      "source": [
        "# create a pivot table to see the numbers of users from the sample grouped by group, language and conversion indicator\n",
        "\n",
        "groups = df.group.unique()\n",
        "pages = ['Existing page', 'New page', 'Total']\n",
        "langs = df.language_preferred.unique()\n",
        "\n",
        "pt = df.pivot_table(index=['group', 'language_preferred'], columns='converted', values='time_spent_on_the_page', aggfunc='count')\n",
        "pt"
      ],
      "metadata": {
        "id": "wugCiB_8wl3h",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 269
        },
        "outputId": "a0e4e371-a26c-45e7-822d-151f92219a02"
      },
      "id": "wugCiB_8wl3h",
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "converted                     no  yes\n",
              "group     language_preferred         \n",
              "control   English              5   11\n",
              "          French              14    3\n",
              "          Spanish             10    7\n",
              "treatment English              6   10\n",
              "          French               5   12\n",
              "          Spanish              6   11"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-46f34731-1ede-452f-8eba-cfe12f362c28\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>converted</th>\n",
              "      <th>no</th>\n",
              "      <th>yes</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>group</th>\n",
              "      <th>language_preferred</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th rowspan=\"3\" valign=\"top\">control</th>\n",
              "      <th>English</th>\n",
              "      <td>5</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>French</th>\n",
              "      <td>14</td>\n",
              "      <td>3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Spanish</th>\n",
              "      <td>10</td>\n",
              "      <td>7</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"3\" valign=\"top\">treatment</th>\n",
              "      <th>English</th>\n",
              "      <td>6</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>French</th>\n",
              "      <td>5</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Spanish</th>\n",
              "      <td>6</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-46f34731-1ede-452f-8eba-cfe12f362c28')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-46f34731-1ede-452f-8eba-cfe12f362c28 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-46f34731-1ede-452f-8eba-cfe12f362c28');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 40
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Pie charts of the proportions\n",
        "fig, ax = plt.subplots(nrows=2, ncols=3, figsize=(12, 8))\n",
        "fig.suptitle('The shares of users visited landing pages', fontsize=18);\n",
        "colors = sns.color_palette('pastel')[0:2]\n",
        "\n",
        "legend_handles = [Patch(color=colors[0]), Patch(color=colors[1])]\n",
        "legend_labels = ['Not converted', 'Converted']\n",
        "\n",
        "for i, group in enumerate(groups):\n",
        "  for j, lang in enumerate(langs):\n",
        "    ax[i, j].pie(pt.loc[(group, lang)], colors=colors, autopct=make_autopct(pt.loc[(group, lang)]), textprops={'fontsize': 12})\n",
        "    ax[i, j].add_artist(plt.Circle((0, 0), 0.25, fc='white'))\n",
        "    if i == 0: ax[i, j].set_title(lang, weight='bold', fontsize=14)\n",
        "    if j == 0: ax[i, j].set_ylabel(pages[i], weight='bold').set_fontsize('14')\n",
        "\n",
        "ax[0,2].legend(handles=legend_handles, labels=legend_labels, loc='upper right', bbox_to_anchor=(1.8, 1.2, 0, 0), fontsize=14)\n",
        "\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 523
        },
        "id": "KVqrMWblvNvJ",
        "outputId": "2c64b2f6-7dfa-45d2-c5a4-87f97f2c3ff0"
      },
      "id": "KVqrMWblvNvJ",
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 864x576 with 6 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {}
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* There are a total of 50 users in the control group (existing page) sample\n",
        "  * In the Spanish language group, 7 out of 17 (or 41%) users were converted\n",
        "  * In the English language group, 11 out of 16 (or 69%) users were converted\n",
        "  * In the French language group, 3 out of 17 (or 18%) users were converted\n",
        "\n",
        "* There are a total of 50 users in the treatment group (new page) sample\n",
        "  * In the Spanish language group, 11 out of 17 (or 65%) users were converted\n",
        "  * In the English language group, 10 out of 16 (or 62%) users were converted\n",
        "  * In the French language group, 12 out of 17 (or 71%) users were converted"
      ],
      "metadata": {
        "id": "pOituWj4fImB"
      },
      "id": "pOituWj4fImB"
    },
    {
      "cell_type": "markdown",
      "source": [
        "To vizaulize the relationship between the two categorical variables, we can plot a stacked bar charts."
      ],
      "metadata": {
        "id": "Uz1ibJPshC1V"
      },
      "id": "Uz1ibJPshC1V"
    },
    {
      "cell_type": "code",
      "source": [
        "# Pie charts of the proportions\n",
        "fig, ax = plt.subplots(nrows=1, ncols=2, figsize=(10, 4))\n",
        "fig.suptitle('Conversion by language and group', fontsize=18);\n",
        "\n",
        "colors = sns.color_palette('pastel')[0:2]\n",
        "dfc=df[df.group=='control']\n",
        "dft=df[df.group=='treatment']\n",
        "\n",
        "pd.crosstab(index=dfc.language_preferred, columns=dfc.converted, normalize=True).plot(kind='bar', stacked=True, color=colors, ax=ax[0])\n",
        "ax[0].set_xlabel('Control group (existing page)', fontsize=14)\n",
        "ax[0].tick_params(axis='x', labelrotation=0)\n",
        "pd.crosstab(index=dft.language_preferred, columns=dft.converted, normalize=True).plot(kind='bar', stacked=True, color=colors, ax=ax[1])\n",
        "ax[1].set_xlabel('Treatment group (new page)', fontsize=14)\n",
        "ax[1].tick_params(axis='x', labelrotation=0)\n",
        "\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 312
        },
        "id": "tD0ucyF4hP-3",
        "outputId": "72617754-445e-48b7-a525-323f95cd7f35"
      },
      "id": "tD0ucyF4hP-3",
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 720x288 with 2 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "* Control group (existing page)\n",
        "  * The proportion of converted users is significantly low in the French language group (compared with two other groups)\n",
        "  * Based on the sample data, we can infer that the proportion of converted users is lower for one language group, but we can not say if this difference is significant enough to conclude that the conversion is dependent on the preferred language\n",
        "\n",
        "* Treatment group (new page)\n",
        "  * The proportion of converted users seems similar for all the language groups\n",
        "  * The share of converted users is greater than not-converted\n",
        "  * Based on the sample data, we can infer that the proportion of converted users is the same for all the language groups\n",
        "\n",
        "* Existing page vs. New page\n",
        "  * Conversion of the English language group is similar for the existing and the new pages\n",
        "  * Conversion of the French language group is significantly higher on the new page\n",
        "  * Conversion of the Spanish language group is slightly higher on the new page\n"
      ],
      "metadata": {
        "id": "ZJQwNM6kmScz"
      },
      "id": "ZJQwNM6kmScz"
    },
    {
      "cell_type": "markdown",
      "source": [
        "###**Hypothesis Testing**"
      ],
      "metadata": {
        "id": "suOYFKL3CR3a"
      },
      "id": "suOYFKL3CR3a"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 1: Define the null and alternate hypotheses"
      ],
      "metadata": {
        "id": "bR5xKCyySsWw"
      },
      "id": "bR5xKCyySsWw"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Let's write the null and alternative hypothesis**\n",
        "\n",
        "We will test the null hypothesis\n",
        "\n",
        ">$H_0:$ Conversion is independent of preferred language.\n",
        "\n",
        "against the alternate hypothesis\n",
        "\n",
        ">$H_a:$ Conversion depends on preferred language.\n"
      ],
      "metadata": {
        "id": "B2qw4Ny1MkYf"
      },
      "id": "B2qw4Ny1MkYf"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 2: Select Appropriate test"
      ],
      "metadata": {
        "id": "BvhygxEfSsWx"
      },
      "id": "BvhygxEfSsWx"
    },
    {
      "cell_type": "markdown",
      "source": [
        "A Chi-square test for independence is used to test the dependence between two categorical variables.\n",
        "\n",
        "A **Chi-square test for independence** is based on the Chi-square probability distribution. It involves the calculation of a chi-square test statistic. A chi-square ($\\chi^2$) statistic is a measure of the difference between the observed and expected frequencies of the outcomes of a set of events or variables, given the size of the sample and the number of variables in the relationship. The test compares what we observe in the random sample to what we expect to observe when we assume there is no relationship between the two variables."
      ],
      "metadata": {
        "id": "0FI8d14As9gU"
      },
      "id": "0FI8d14As9gU"
    },
    {
      "cell_type": "code",
      "source": [
        "# Chech the number of sample observations in each level of the variable \n",
        "\n",
        "print('Numbers of sample observations in each level of the variable\\n','-'*60)\n",
        "df.pivot_table(index='language_preferred', columns='converted', values='time_spent_on_the_page', aggfunc='count', margins=True)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 241
        },
        "id": "TfIxh4ZONUpl",
        "outputId": "6891b81a-8bba-46ee-f0e0-e8bba5523cd1"
      },
      "id": "TfIxh4ZONUpl",
      "execution_count": 43,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Numbers of sample observations in each level of the variable\n",
            " ------------------------------------------------------------\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "converted           no  yes  All\n",
              "language_preferred              \n",
              "English             11   21   32\n",
              "French              19   15   34\n",
              "Spanish             16   18   34\n",
              "All                 46   54  100"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-3c8adb6f-f1d3-4981-953a-9d8bc099693f\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>converted</th>\n",
              "      <th>no</th>\n",
              "      <th>yes</th>\n",
              "      <th>All</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>language_preferred</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>English</th>\n",
              "      <td>11</td>\n",
              "      <td>21</td>\n",
              "      <td>32</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>French</th>\n",
              "      <td>19</td>\n",
              "      <td>15</td>\n",
              "      <td>34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Spanish</th>\n",
              "      <td>16</td>\n",
              "      <td>18</td>\n",
              "      <td>34</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>All</th>\n",
              "      <td>46</td>\n",
              "      <td>54</td>\n",
              "      <td>100</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-3c8adb6f-f1d3-4981-953a-9d8bc099693f')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-3c8adb6f-f1d3-4981-953a-9d8bc099693f button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-3c8adb6f-f1d3-4981-953a-9d8bc099693f');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 43
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Let's test whether the assumptions are satisfied or not**\n",
        "\n",
        "- [x] Categorical variables, **Yes**, `converted` and `language_preferred` are categorical variables\n",
        "- [x] Expected value of the number of sample observations in each level of the variable is at least 5 - **Yes**, the number of observations in each level is greater than 5\n",
        "- [x] Random sampling from the population - **Yes**, we are informed that the collected sample is a simple random sample"
      ],
      "metadata": {
        "id": "eVNecWF1M5WB"
      },
      "id": "eVNecWF1M5WB"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 3: Decide the significance level"
      ],
      "metadata": {
        "id": "DVNdmtUSSsWx"
      },
      "id": "DVNdmtUSSsWx"
    },
    {
      "cell_type": "markdown",
      "source": [
        "As given in the problem statement, we select $\\alpha$ as $0.05$."
      ],
      "metadata": {
        "id": "Mg1cjv0cRa1f"
      },
      "id": "Mg1cjv0cRa1f"
    },
    {
      "cell_type": "code",
      "source": [
        "alpha = 0.05"
      ],
      "metadata": {
        "id": "VzuhBdSbRa1f"
      },
      "execution_count": 44,
      "outputs": [],
      "id": "VzuhBdSbRa1f"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 4: Collect and prepare data"
      ],
      "metadata": {
        "id": "Or4nvtg1SsWx"
      },
      "id": "Or4nvtg1SsWx"
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let's prepare the data for the test. First, to perform the chi-square test of independence, we need to create the contingency table, for which we will use pd.crosstab()"
      ],
      "metadata": {
        "id": "snAh6mnkxm6R"
      },
      "id": "snAh6mnkxm6R"
    },
    {
      "cell_type": "code",
      "source": [
        "# Treatment group (new page)\n",
        "contingency_table = pd.crosstab(dft.language_preferred, dft.converted)\n",
        "contingency_table"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 175
        },
        "id": "7JPrIcvRRe7y",
        "outputId": "cfb6aed3-4b8b-4f73-eb0a-0d9cfaa803fa"
      },
      "id": "7JPrIcvRRe7y",
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "converted           no  yes\n",
              "language_preferred         \n",
              "English              6   10\n",
              "French               5   12\n",
              "Spanish              6   11"
            ],
            "text/html": [
              "\n",
              "  <div id=\"df-a414c2f0-9126-483a-9a0b-7a5250c9b5d8\">\n",
              "    <div class=\"colab-df-container\">\n",
              "      <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>converted</th>\n",
              "      <th>no</th>\n",
              "      <th>yes</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>language_preferred</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>English</th>\n",
              "      <td>6</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>French</th>\n",
              "      <td>5</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Spanish</th>\n",
              "      <td>6</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>\n",
              "      <button class=\"colab-df-convert\" onclick=\"convertToInteractive('df-a414c2f0-9126-483a-9a0b-7a5250c9b5d8')\"\n",
              "              title=\"Convert this dataframe to an interactive table.\"\n",
              "              style=\"display:none;\">\n",
              "        \n",
              "  <svg xmlns=\"http://www.w3.org/2000/svg\" height=\"24px\"viewBox=\"0 0 24 24\"\n",
              "       width=\"24px\">\n",
              "    <path d=\"M0 0h24v24H0V0z\" fill=\"none\"/>\n",
              "    <path d=\"M18.56 5.44l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94zm-11 1L8.5 8.5l.94-2.06 2.06-.94-2.06-.94L8.5 2.5l-.94 2.06-2.06.94zm10 10l.94 2.06.94-2.06 2.06-.94-2.06-.94-.94-2.06-.94 2.06-2.06.94z\"/><path d=\"M17.41 7.96l-1.37-1.37c-.4-.4-.92-.59-1.43-.59-.52 0-1.04.2-1.43.59L10.3 9.45l-7.72 7.72c-.78.78-.78 2.05 0 2.83L4 21.41c.39.39.9.59 1.41.59.51 0 1.02-.2 1.41-.59l7.78-7.78 2.81-2.81c.8-.78.8-2.07 0-2.86zM5.41 20L4 18.59l7.72-7.72 1.47 1.35L5.41 20z\"/>\n",
              "  </svg>\n",
              "      </button>\n",
              "      \n",
              "  <style>\n",
              "    .colab-df-container {\n",
              "      display:flex;\n",
              "      flex-wrap:wrap;\n",
              "      gap: 12px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert {\n",
              "      background-color: #E8F0FE;\n",
              "      border: none;\n",
              "      border-radius: 50%;\n",
              "      cursor: pointer;\n",
              "      display: none;\n",
              "      fill: #1967D2;\n",
              "      height: 32px;\n",
              "      padding: 0 0 0 0;\n",
              "      width: 32px;\n",
              "    }\n",
              "\n",
              "    .colab-df-convert:hover {\n",
              "      background-color: #E2EBFA;\n",
              "      box-shadow: 0px 1px 2px rgba(60, 64, 67, 0.3), 0px 1px 3px 1px rgba(60, 64, 67, 0.15);\n",
              "      fill: #174EA6;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert {\n",
              "      background-color: #3B4455;\n",
              "      fill: #D2E3FC;\n",
              "    }\n",
              "\n",
              "    [theme=dark] .colab-df-convert:hover {\n",
              "      background-color: #434B5C;\n",
              "      box-shadow: 0px 1px 3px 1px rgba(0, 0, 0, 0.15);\n",
              "      filter: drop-shadow(0px 1px 2px rgba(0, 0, 0, 0.3));\n",
              "      fill: #FFFFFF;\n",
              "    }\n",
              "  </style>\n",
              "\n",
              "      <script>\n",
              "        const buttonEl =\n",
              "          document.querySelector('#df-a414c2f0-9126-483a-9a0b-7a5250c9b5d8 button.colab-df-convert');\n",
              "        buttonEl.style.display =\n",
              "          google.colab.kernel.accessAllowed ? 'block' : 'none';\n",
              "\n",
              "        async function convertToInteractive(key) {\n",
              "          const element = document.querySelector('#df-a414c2f0-9126-483a-9a0b-7a5250c9b5d8');\n",
              "          const dataTable =\n",
              "            await google.colab.kernel.invokeFunction('convertToInteractive',\n",
              "                                                     [key], {});\n",
              "          if (!dataTable) return;\n",
              "\n",
              "          const docLinkHtml = 'Like what you see? Visit the ' +\n",
              "            '<a target=\"_blank\" href=https://colab.research.google.com/notebooks/data_table.ipynb>data table notebook</a>'\n",
              "            + ' to learn more about interactive tables.';\n",
              "          element.innerHTML = '';\n",
              "          dataTable['output_type'] = 'display_data';\n",
              "          await google.colab.output.renderOutput(dataTable, element);\n",
              "          const docLink = document.createElement('div');\n",
              "          docLink.innerHTML = docLinkHtml;\n",
              "          element.appendChild(docLink);\n",
              "        }\n",
              "      </script>\n",
              "    </div>\n",
              "  </div>\n",
              "  "
            ]
          },
          "metadata": {},
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 5: Calculate the p-value"
      ],
      "metadata": {
        "id": "zcT3P_TGSsWx"
      },
      "id": "zcT3P_TGSsWx"
    },
    {
      "cell_type": "markdown",
      "source": [
        "We will use the [`chi2_contingency()`](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.chi2_contingency.html) function from the Scipy stats library to perform the hypothesis test. It takes the contingency table as an input and returns the test statistic, p-value, degrees of freedom, and the expected frequencies as the output."
      ],
      "metadata": {
        "id": "fQMyApidyvxX"
      },
      "id": "fQMyApidyvxX"
    },
    {
      "cell_type": "code",
      "source": [
        "# find the p-value\n",
        "chi, p_value, dof, expected = chi2_contingency(contingency_table)\n",
        "print('The p-value is', p_value)"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GcAAVKZWRSGZ",
        "outputId": "6fd22e23-a851-47ab-80e6-cdcd7bf35ab5"
      },
      "id": "GcAAVKZWRSGZ",
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.8783081441552572\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 6: Compare the p-value with $\\alpha$"
      ],
      "metadata": {
        "id": "9LN6UymMSsWx"
      },
      "id": "9LN6UymMSsWx"
    },
    {
      "cell_type": "code",
      "source": [
        "if p_value > alpha:\n",
        "  print(f'As the p-value ({p_value:.3f}) is greater than the significance level ({alpha}), we fail to reject the null hypothesis')\n",
        "else:\n",
        "  print(f'As the p-value ({p_value:.3f}) is less than (or equal to) the significance level ({alpha}), we can reject the null hypothesis')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wF1r1KAlSM-j",
        "outputId": "9aad8b5e-9c0d-4a2f-d06f-99196d4a65fa"
      },
      "id": "wF1r1KAlSM-j",
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "As the p-value (0.878) is greater than the significance level (0.05), we fail to reject the null hypothesis\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 6.1: Calculate p-value and compare the p-value with $\\alpha$ for the control group (existing page)"
      ],
      "metadata": {
        "id": "XqgQSgxzW6m0"
      },
      "id": "XqgQSgxzW6m0"
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "b0e63f17",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "bda7d998-15c4-4a58-a4f0-1d4e8639880e"
      },
      "source": [
        "# Control group (old page)\n",
        "contingency_table = pd.crosstab(dfc.language_preferred, dfc.converted)\n",
        "chi, p_value, dof, expected = chi2_contingency(contingency_table)\n",
        "if p_value > alpha: print(f'As the p-value ({p_value:.3f}) is greater than the significance level ({alpha}), we fail to reject the null hypothesis')\n",
        "else: print(f'As the p-value ({p_value:.3f}) is less than the significance level ({alpha}), we reject the null hypothesis')"
      ],
      "id": "b0e63f17",
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "As the p-value (0.012) is less than the significance level (0.05), we reject the null hypothesis\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 7:  Draw inference"
      ],
      "metadata": {
        "id": "1ggaoyI9SsWx"
      },
      "id": "1ggaoyI9SsWx"
    },
    {
      "cell_type": "markdown",
      "source": [
        ">Since the p-value is greater than the 5% significance level, we fail to reject the null hypothesis. Hence, we do not have enough statistical evidence to say that the conversion depends on the preferred language for the treatment group of users of the new landing page.\n",
        ">  * We reject the null hypothesis for users of the existing page, as the p-value is less than the 5% significance level. Hence, we have enough statistical significance to say that conversion depends on the preferred language for the control group of users who used the existing landing page.\n",
        ">  * It seems that the new page was significantly improved for French language users from the conversion standpoint<br>\n",
        ">\n",
        ">**The conversion and preferred language are independent** for the treatment group (new page)"
      ],
      "metadata": {
        "id": "BU9gGtFpSYmX"
      },
      "id": "BU9gGtFpSYmX"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "8d585a90"
      },
      "source": [
        "##**4. Is the time spent on the new page same for the different language users?**"
      ],
      "id": "8d585a90"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Perform Visual Analysis"
      ],
      "metadata": {
        "id": "3hdb3FIfSubX"
      },
      "id": "3hdb3FIfSubX"
    },
    {
      "cell_type": "code",
      "source": [
        "new = df[df['group']=='treatment']\n",
        "new.groupby('language_preferred').mean()['time_spent_on_the_page']"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "27WcugEQl5uU",
        "outputId": "8a20a087-78e7-446b-cb2e-a71d71cee813"
      },
      "id": "27WcugEQl5uU",
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "language_preferred\n",
              "English    6.663750\n",
              "French     6.196471\n",
              "Spanish    5.835294\n",
              "Name: time_spent_on_the_page, dtype: float64"
            ]
          },
          "metadata": {},
          "execution_count": 49
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Boxplot of `time_spent_on_the_page` by languages\n",
        "fig, ax = plt.subplots(figsize=(6, 6)) \n",
        "sns.boxplot(y=new['time_spent_on_the_page'], x=new['language_preferred'])\n",
        "ax.set_title('Time spent on the new page for the different language users')\n",
        "ax.set_ylabel('Time spent (min)', fontsize=14)\n",
        "ax.set_xlabel('Language', fontsize=14)\n",
        "\n",
        "plt.tight_layout()\n",
        "plt.show();"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 441
        },
        "id": "heJqD-c4mtZp",
        "outputId": "742e75c9-911e-4765-a9c1-58d1053ce058"
      },
      "id": "heJqD-c4mtZp",
      "execution_count": 50,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "text/plain": [
              "<Figure size 432x432 with 1 Axes>"
            ],
            "image/png": "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\n"
          },
          "metadata": {
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Here, `time_spent_on_the_page` is the response and `language_preferred` is the factor.**"
      ],
      "metadata": {
        "id": "JuVTupnRsPdH"
      },
      "id": "JuVTupnRsPdH"
    },
    {
      "cell_type": "markdown",
      "source": [
        "* The distribution of time spent seem to differ among the three groups\n",
        "* English seems to impact the longest time spent on the new landing page\n",
        "* The median time spent seems to be very close for English and Spanish preferred language groups, but the variation is less for Spanish as compared to English and French\n",
        "\n",
        "*We cannot say that the observed difference in time spent among the three groups is significant enough to conclude the same about the three language groups.*\n",
        "\n",
        "**To determine this, we will test the difference using a statistical test.**"
      ],
      "metadata": {
        "id": "hXQx3q90uMOk"
      },
      "id": "hXQx3q90uMOk"
    },
    {
      "cell_type": "markdown",
      "source": [
        "###**Hypothesis Testing**"
      ],
      "metadata": {
        "id": "6HcReMnRCUYt"
      },
      "id": "6HcReMnRCUYt"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 1: Define the null and alternate hypotheses"
      ],
      "metadata": {
        "id": "1R84kXn4SubY"
      },
      "id": "1R84kXn4SubY"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Let's write the null and alternative hypothesis**\n",
        "\n",
        "The null and alternative hypotheses can be formulated as:\n",
        "\n",
        "> $H_0$ : The mean time spent on the new page with respect to each language preferences is equal. <br>\n",
        "> $H_a$ : At least one of the mean time spent on the new page with respect to the three language preferences is different.\n",
        "\n",
        "Let $\\mu_1, \\mu_2, \\mu_3$ be the means of the time spent on the new page for the English, Spanish, and French users, respectively. We will test the null hypothesis.\n",
        "\n",
        "Mathematically, the above-formulated hypotheses can be written as follows:\n",
        "\n",
        ">$H_0: \\mu_1 = \\mu_2 = \\mu_3$ <br>\n",
        ">$H_a:$ At least one of the means is different from the rest.\n"
      ],
      "metadata": {
        "id": "FMUkaStHn5TJ"
      },
      "id": "FMUkaStHn5TJ"
    },
    {
      "cell_type": "markdown",
      "source": [
        "Let us test the normality of the distribution of time users spent on the new landing page and the homogeneity of variance for three different language groups."
      ],
      "metadata": {
        "id": "GpQzBH5eFYLG"
      },
      "id": "GpQzBH5eFYLG"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Shapiro-Wilk’s test (Assumption 1: Normality)**\n",
        "\n",
        "We will test the null hypothesis\n",
        "\n",
        ">$H_0:$ The time spent on the new page follows a normal distribution\n",
        "\n",
        "against the alternative hypothesis\n",
        "\n",
        ">$H_a:$ The time spent on the new page does not follow a normal distribution"
      ],
      "metadata": {
        "id": "PGScDzwjp5at"
      },
      "id": "PGScDzwjp5at"
    },
    {
      "cell_type": "code",
      "source": [
        "# The shapiro() function of Scipy will be used to compute the test statistic and p-value\n",
        "\n",
        "# find the p-value\n",
        "w, p_value = stats.shapiro(new['time_spent_on_the_page']) \n",
        "print(f'The p-value is {p_value:.2f}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "pbBnZ7CUp6Yn",
        "outputId": "dd9dbebd-5a37-4fd3-c967-c2a99376c1ca"
      },
      "id": "pbBnZ7CUp6Yn",
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.80\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        ">Since p-value of the test is very large, we fail to reject the null hypothesis that the response (the time users spent on the new page) follows the normal distribution."
      ],
      "metadata": {
        "id": "l2-6pJH7rCSF"
      },
      "id": "l2-6pJH7rCSF"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**Levene’s test (Assumption 2: Homogeneity of Variance)**\n",
        "\n",
        "We will test the null hypothesis.\n",
        "\n",
        ">$H_0$: All the variances are equal\n",
        "\n",
        "against the alternative hypothesis\n",
        "\n",
        ">$H_a$: At least one variance is different from the rest"
      ],
      "metadata": {
        "id": "jXXeaJOzrbjB"
      },
      "id": "jXXeaJOzrbjB"
    },
    {
      "cell_type": "code",
      "source": [
        "# The levene() function of Scipy will be used to compute the test statistic and p-value\n",
        "\n",
        "statistic, p_value = stats.levene(new['time_spent_on_the_page'][new['language_preferred']==\"English\"],\n",
        "                                  new['time_spent_on_the_page'][new['language_preferred']==\"Spanish\"],\n",
        "                                  new['time_spent_on_the_page'][new['language_preferred']==\"French\"])\n",
        "# find the p-value\n",
        "print(f'The p-value is {p_value:.2f}')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "hPbqLDt8rqGC",
        "outputId": "47fd6f62-5f05-437e-987e-f31bf55edebd"
      },
      "id": "hPbqLDt8rqGC",
      "execution_count": 52,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.47\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        ">Since the p-value is large, we fail to reject the null hypothesis of homogeneity of variances"
      ],
      "metadata": {
        "id": "cN9wGiFrrpTh"
      },
      "id": "cN9wGiFrrpTh"
    },
    {
      "cell_type": "markdown",
      "source": [
        "**The normality and equality of variance assumptions checks**\n",
        "\n",
        "- [x] For testing of normality, Shapiro-Wilk’s test is applied to the response variable.\n",
        "- [x] For equality of variance, Levene test is applied to the response variable.\n"
      ],
      "metadata": {
        "id": "qoW0LXIkpjvG"
      },
      "id": "qoW0LXIkpjvG"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 2: Select Appropriate test"
      ],
      "metadata": {
        "id": "F6hknohwSubY"
      },
      "id": "F6hknohwSubY"
    },
    {
      "cell_type": "markdown",
      "source": [
        "This is a problem, concerning three population means. **One-way ANOVA** is an appropriate test here provided normality and equality of variance assumptions are verified.\n",
        "\n",
        "**One-way ANOVA test**\n",
        "\n",
        "In a one-way ANOVA test, we compare the means from several populations to test if there is any significance difference between them. The results from an ANOVA test are most reliable when the assumptions of normality and equality of variances are satisfied.\n",
        "\n",
        "**Let's test whether the assumptions are satisfied or not**\n",
        "\n",
        "- [x] The populations are normally distributed - **Yes**, the normality assumption is verified using the Shapiro-Wilk’s test\n",
        "- [x] Samples are independent simple random samples - **Yes**, we are informed that the collected sample is a simple random sample\n",
        "- [x] Population variances are equal - **Yes**, the homogeneity of variance assumption is verified using the Levene's test"
      ],
      "metadata": {
        "id": "inNEvZDOtyB2"
      },
      "id": "inNEvZDOtyB2"
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 3: Decide the significance level"
      ],
      "metadata": {
        "id": "-2KV1mTUSubZ"
      },
      "id": "-2KV1mTUSubZ"
    },
    {
      "cell_type": "markdown",
      "source": [
        "As given in the problem statement, we select $\\alpha$ as $0.05$."
      ],
      "metadata": {
        "id": "8g2iNisSxfe9"
      },
      "id": "8g2iNisSxfe9"
    },
    {
      "cell_type": "code",
      "source": [
        "alpha = 0.05"
      ],
      "metadata": {
        "id": "4jZwYLmoxeiL"
      },
      "id": "4jZwYLmoxeiL",
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 4: Collect and prepare data"
      ],
      "metadata": {
        "id": "tQSKq-RCSubZ"
      },
      "id": "tQSKq-RCSubZ"
    },
    {
      "cell_type": "code",
      "source": [
        "# create separate variables to store the time spent with respect to the three languages\n",
        "en = new[new['language_preferred']==\"English\"]['time_spent_on_the_page']\n",
        "es = new[new['language_preferred']==\"Spanish\"]['time_spent_on_the_page']\n",
        "fr = new[new['language_preferred']==\"French\"]['time_spent_on_the_page']"
      ],
      "metadata": {
        "id": "UMsf0FdkxqCJ"
      },
      "id": "UMsf0FdkxqCJ",
      "execution_count": 54,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 5: Calculate the p-value"
      ],
      "metadata": {
        "id": "hHltPjl5SubZ"
      },
      "id": "hHltPjl5SubZ"
    },
    {
      "cell_type": "markdown",
      "source": [
        "* We will use the [`f_oneway()`](https://docs.scipy.org/doc/scipy/reference/generated/scipy.stats.f_oneway.html]) function from the `scipy.stats` library to perform a one-way ANOVA test.\n",
        "* The `f_oneway()` function takes the sample observations from the different groups and returns:\n",
        "  * the test statistic and \n",
        "  * the p-value for the test.\n",
        "* The sample observations are the values of time spent with respect to the three language preferences."
      ],
      "metadata": {
        "id": "QcYIbjbFyN9A"
      },
      "id": "QcYIbjbFyN9A"
    },
    {
      "cell_type": "code",
      "source": [
        "# find the p-value\n",
        "test_stat, p_value = stats.f_oneway(en, es, fr)\n",
        "print(f'The p-value is {p_value:.4f}')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ygAp1_BjyLTr",
        "outputId": "203e1700-b477-4903-dfa3-7b774c288d22"
      },
      "id": "ygAp1_BjyLTr",
      "execution_count": 55,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "The p-value is 0.4320\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 6: Compare the p-value with $\\alpha$"
      ],
      "metadata": {
        "id": "1vgEgwIiSubZ"
      },
      "id": "1vgEgwIiSubZ"
    },
    {
      "cell_type": "code",
      "source": [
        "# print the conclusion based on p-value\n",
        "if p_value < alpha:\n",
        "    print(f'As the p-value {p_value:.5f} is less than the level of significance, we reject the null hypothesis.')\n",
        "else:\n",
        "    print(f'As the p-value {p_value:.5f} is greater than the level of significance, we fail to reject the null hypothesis.')"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "O6X2Kqr0zB-d",
        "outputId": "7b3c2460-8d1c-4706-ccb0-4c447886ede3"
      },
      "id": "O6X2Kqr0zB-d",
      "execution_count": 56,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "As the p-value 0.43204 is greater than the level of significance, we fail to reject the null hypothesis.\n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "### Step 7:  Draw inference"
      ],
      "metadata": {
        "id": "z06wOwy7SubZ"
      },
      "id": "z06wOwy7SubZ"
    },
    {
      "cell_type": "markdown",
      "source": [
        ">Since the p-value is greater than the level of significance (5%), we fail to reject the null hypothesis. <br>\n",
        ">Hence, we do not have enough statistical evidence to say that at least one of the means of time spent by different language preference users on the new landing page differs from others.<br>\n",
        ">\n",
        ">**The time spent on the new page is the same for the different language users** \n"
      ],
      "metadata": {
        "id": "LY9go0FzzIt-"
      },
      "id": "LY9go0FzzIt-"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3acbc947"
      },
      "source": [
        "##**Conclusion and Business Recommendations**"
      ],
      "id": "3acbc947"
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "21fb1063"
      },
      "source": [
        "* **Answers to the questions**\n",
        "  * (1) We have enough statistical evidence to say that the average time users spend on the new landing page is significantly greater than the average time spent on the existing landing page.\n",
        "  * (2) We have enough statistical evidence to say that the conversion rate (the proportion of users who visit the landing page and get converted) for the new page is greater than the conversion rate for the old page.\n",
        "  * (3) We do not have enough statistical evidence to say that the conversion depends on the preferred language for the treatment group of users of the new landing page.\n",
        "  * (4) We do not have enough statistical evidence to say that at least one of the means of time spent by different language preference users on the new landing page differs from others.\n",
        "\n",
        "* **Conclusions**\n",
        "  * The users spend more time on the new landing page than on the existing landing page.\n",
        "  * The conversion rate for the new page is greater than the conversion rate for the old page.\n",
        "  * The conversion and preferred language are independent in the treatment group (new page) but are dependent in the control group (existing page)\n",
        "  * The time spent on the new page is the same for the different language users\n",
        "  * The variations in the times users spent on the new landing page is less than on the existing page\n",
        "  * The conversion rate is 1.55 times higher on the new landing page was happened mainly due to a significant increase in the conversion rate for the French language preference\n",
        "  * The conversion rate does not depend on the preferred language. The conversion rate for the French language on the new page has been significantly improved compared to the existing landing page.\n",
        "\n",
        "* **Business Recommendations**\n",
        "  * The new landing page implementation may increase the average time users spend on the landing page and the conversion rate for all languages.\n",
        "  * The new landing page rollout is recommended as it fixes some issues on the existing page with the French language preference and avoids a low conversion rate for this option.\n"
      ],
      "id": "21fb1063"
    },
    {
      "cell_type": "markdown",
      "source": [
        "___"
      ],
      "metadata": {
        "id": "U0dAd4i0iCHo"
      },
      "id": "U0dAd4i0iCHo"
    },
    {
      "cell_type": "code",
      "source": [
        "%%shell\n",
        "\n",
        "jupyter nbconvert --to html /content/project_2_at.ipynb"
      ],
      "metadata": {
        "id": "yRCDBNf0Wlv2"
      },
      "id": "yRCDBNf0Wlv2",
      "execution_count": 56,
      "outputs": []
    }
  ]
}