{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "83e16a20",
   "metadata": {},
   "source": [
    "# Demo: Using `create_odds_ratios` from the vivainsights Python Package\n",
    "\n",
    "This notebook demonstrates how to use the `create_odds_ratios` function from the **vivainsights** Python package to analyze the relationship between ordinal metrics and an independent variable.\n",
    "\n",
    "In this walkthrough, you will:\n",
    "1. Load demo data (`pq_data`) from the package.\n",
    "2. Create an independent variable (`UsageSegments_12w`) using `identify_usage_segments`.\n",
    "3. Compute favorability scores for ordinal metrics with `compute_fav`.\n",
    "4. Calculate odds ratios for ordinal metrics using `create_odds_ratios`.\n",
    "5. Visualize the results for easier interpretation."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "14bb79c7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import necessary libraries\n",
    "import vivainsights as vi\n",
    "import pandas as pd\n",
    "import warnings\n",
    "\n",
    "# Suppress warnings for cleaner output\n",
    "warnings.filterwarnings(\"ignore\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a9cc5db8",
   "metadata": {},
   "source": [
    "## Step 1: Load the demo data\n",
    "\n",
    "First, load the sample Person Query dataset (`pq_data`) provided by **vivainsights**."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "53148e9d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PersonId</th>\n",
       "      <th>MetricDate</th>\n",
       "      <th>Collaboration_hours</th>\n",
       "      <th>Copilot_actions_taken_in_Teams</th>\n",
       "      <th>Meeting_and_call_hours</th>\n",
       "      <th>Internal_network_size</th>\n",
       "      <th>Email_hours</th>\n",
       "      <th>Channel_message_posts</th>\n",
       "      <th>Conflicting_meeting_hours</th>\n",
       "      <th>Large_and_long_meeting_hours</th>\n",
       "      <th>...</th>\n",
       "      <th>Summarise_chat_actions_taken_using_Copilot_in_Teams</th>\n",
       "      <th>Summarise_email_thread_actions_taken_using_Copilot_in_Outlook</th>\n",
       "      <th>Summarise_meeting_actions_taken_using_Copilot_in_Teams</th>\n",
       "      <th>Summarise_presentation_actions_taken_using_Copilot_in_PowerPoint</th>\n",
       "      <th>Summarise_Word_document_actions_taken_using_Copilot_in_Word</th>\n",
       "      <th>FunctionType</th>\n",
       "      <th>SupervisorIndicator</th>\n",
       "      <th>Level</th>\n",
       "      <th>Organization</th>\n",
       "      <th>LevelDesignation</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>bf361ad4-fc29-432f-95f3-837e689f4ac4</td>\n",
       "      <td>2024-03-31</td>\n",
       "      <td>17.452987</td>\n",
       "      <td>4</td>\n",
       "      <td>11.767599</td>\n",
       "      <td>92</td>\n",
       "      <td>7.523189</td>\n",
       "      <td>0.753451</td>\n",
       "      <td>2.079210</td>\n",
       "      <td>0.635489</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Specialist</td>\n",
       "      <td>Manager</td>\n",
       "      <td>Level3</td>\n",
       "      <td>IT</td>\n",
       "      <td>Senior IC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0500f22c-2910-4154-b6e2-66864898d848</td>\n",
       "      <td>2024-03-31</td>\n",
       "      <td>32.860820</td>\n",
       "      <td>6</td>\n",
       "      <td>26.743370</td>\n",
       "      <td>193</td>\n",
       "      <td>11.578396</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.106997</td>\n",
       "      <td>1.402567</td>\n",
       "      <td>...</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Specialist</td>\n",
       "      <td>Manager</td>\n",
       "      <td>Level2</td>\n",
       "      <td>Legal</td>\n",
       "      <td>Senior Manager</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>bb495ec9-8577-468a-8b48-e32677442f51</td>\n",
       "      <td>2024-03-31</td>\n",
       "      <td>21.502359</td>\n",
       "      <td>8</td>\n",
       "      <td>13.982031</td>\n",
       "      <td>113</td>\n",
       "      <td>9.073214</td>\n",
       "      <td>0.894786</td>\n",
       "      <td>3.001401</td>\n",
       "      <td>0.000192</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Manager</td>\n",
       "      <td>Manager</td>\n",
       "      <td>Level4</td>\n",
       "      <td>Legal</td>\n",
       "      <td>Junior IC</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>f6d58aaf-a2b2-42ab-868f-d7ac2e99788d</td>\n",
       "      <td>2024-03-31</td>\n",
       "      <td>25.416502</td>\n",
       "      <td>4</td>\n",
       "      <td>16.895513</td>\n",
       "      <td>131</td>\n",
       "      <td>10.281204</td>\n",
       "      <td>0.528731</td>\n",
       "      <td>1.846423</td>\n",
       "      <td>1.441596</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Manager</td>\n",
       "      <td>Manager</td>\n",
       "      <td>Level1</td>\n",
       "      <td>HR</td>\n",
       "      <td>Executive</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>c81cb49a-aa27-4cfc-8211-4087b733a3c6</td>\n",
       "      <td>2024-03-31</td>\n",
       "      <td>11.433377</td>\n",
       "      <td>4</td>\n",
       "      <td>6.957468</td>\n",
       "      <td>75</td>\n",
       "      <td>5.510535</td>\n",
       "      <td>2.288934</td>\n",
       "      <td>0.474048</td>\n",
       "      <td>0.269996</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Technician</td>\n",
       "      <td>Manager</td>\n",
       "      <td>Level1</td>\n",
       "      <td>Finance</td>\n",
       "      <td>Executive</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 73 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                               PersonId  MetricDate  Collaboration_hours  \\\n",
       "0  bf361ad4-fc29-432f-95f3-837e689f4ac4  2024-03-31            17.452987   \n",
       "1  0500f22c-2910-4154-b6e2-66864898d848  2024-03-31            32.860820   \n",
       "2  bb495ec9-8577-468a-8b48-e32677442f51  2024-03-31            21.502359   \n",
       "3  f6d58aaf-a2b2-42ab-868f-d7ac2e99788d  2024-03-31            25.416502   \n",
       "4  c81cb49a-aa27-4cfc-8211-4087b733a3c6  2024-03-31            11.433377   \n",
       "\n",
       "   Copilot_actions_taken_in_Teams  Meeting_and_call_hours  \\\n",
       "0                               4               11.767599   \n",
       "1                               6               26.743370   \n",
       "2                               8               13.982031   \n",
       "3                               4               16.895513   \n",
       "4                               4                6.957468   \n",
       "\n",
       "   Internal_network_size  Email_hours  Channel_message_posts  \\\n",
       "0                     92     7.523189               0.753451   \n",
       "1                    193    11.578396               0.000000   \n",
       "2                    113     9.073214               0.894786   \n",
       "3                    131    10.281204               0.528731   \n",
       "4                     75     5.510535               2.288934   \n",
       "\n",
       "   Conflicting_meeting_hours  Large_and_long_meeting_hours  ...  \\\n",
       "0                   2.079210                      0.635489  ...   \n",
       "1                   8.106997                      1.402567  ...   \n",
       "2                   3.001401                      0.000192  ...   \n",
       "3                   1.846423                      1.441596  ...   \n",
       "4                   0.474048                      0.269996  ...   \n",
       "\n",
       "   Summarise_chat_actions_taken_using_Copilot_in_Teams  \\\n",
       "0                                                  2     \n",
       "1                                                  2     \n",
       "2                                                  1     \n",
       "3                                                  0     \n",
       "4                                                  0     \n",
       "\n",
       "   Summarise_email_thread_actions_taken_using_Copilot_in_Outlook  \\\n",
       "0                                                  0               \n",
       "1                                                  0               \n",
       "2                                                  1               \n",
       "3                                                  0               \n",
       "4                                                  0               \n",
       "\n",
       "   Summarise_meeting_actions_taken_using_Copilot_in_Teams  \\\n",
       "0                                                  0        \n",
       "1                                                  4        \n",
       "2                                                  0        \n",
       "3                                                  0        \n",
       "4                                                  1        \n",
       "\n",
       "   Summarise_presentation_actions_taken_using_Copilot_in_PowerPoint  \\\n",
       "0                                                  0                  \n",
       "1                                                  1                  \n",
       "2                                                  0                  \n",
       "3                                                  0                  \n",
       "4                                                  0                  \n",
       "\n",
       "   Summarise_Word_document_actions_taken_using_Copilot_in_Word  FunctionType  \\\n",
       "0                                                  0              Specialist   \n",
       "1                                                  0              Specialist   \n",
       "2                                                  0                 Manager   \n",
       "3                                                  0                 Manager   \n",
       "4                                                  0              Technician   \n",
       "\n",
       "   SupervisorIndicator   Level  Organization  LevelDesignation  \n",
       "0              Manager  Level3            IT         Senior IC  \n",
       "1              Manager  Level2         Legal    Senior Manager  \n",
       "2              Manager  Level4         Legal         Junior IC  \n",
       "3              Manager  Level1            HR         Executive  \n",
       "4              Manager  Level1       Finance         Executive  \n",
       "\n",
       "[5 rows x 73 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load the demo data\n",
    "pq_data = vi.load_pq_data()\n",
    "\n",
    "# Display the first few rows of the dataset\n",
    "pq_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "9b6e2684",
   "metadata": {},
   "source": [
    "## Step 2: Create the independent variable with `identify_usage_segments`\n",
    "\n",
    "Use `identify_usage_segments` to classify users into usage segments based on their Copilot actions. The independent variable (`UsageSegments_12w`) is created by aggregating columns that start with `Copilot_actions_taken_in_`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "6e77dc58",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PersonId</th>\n",
       "      <th>MetricDate</th>\n",
       "      <th>Collaboration_hours</th>\n",
       "      <th>Copilot_actions_taken_in_Teams</th>\n",
       "      <th>Meeting_and_call_hours</th>\n",
       "      <th>Internal_network_size</th>\n",
       "      <th>Email_hours</th>\n",
       "      <th>Channel_message_posts</th>\n",
       "      <th>Conflicting_meeting_hours</th>\n",
       "      <th>Large_and_long_meeting_hours</th>\n",
       "      <th>...</th>\n",
       "      <th>Level</th>\n",
       "      <th>Organization</th>\n",
       "      <th>LevelDesignation</th>\n",
       "      <th>target_metric</th>\n",
       "      <th>target_metric_l12w</th>\n",
       "      <th>target_metric_l4w</th>\n",
       "      <th>IsHabit12w</th>\n",
       "      <th>IsHabit4w</th>\n",
       "      <th>UsageSegments_12w</th>\n",
       "      <th>UsageSegments_4w</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>01986072-719a-404c-ae98-009d92e82323</td>\n",
       "      <td>2024-03-31</td>\n",
       "      <td>26.884733</td>\n",
       "      <td>7</td>\n",
       "      <td>17.700027</td>\n",
       "      <td>156</td>\n",
       "      <td>9.667004</td>\n",
       "      <td>0.117751</td>\n",
       "      <td>2.674868</td>\n",
       "      <td>1.262361</td>\n",
       "      <td>...</td>\n",
       "      <td>Level4</td>\n",
       "      <td>IT</td>\n",
       "      <td>Junior IC</td>\n",
       "      <td>10</td>\n",
       "      <td>10.00</td>\n",
       "      <td>10.00</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>Novice User</td>\n",
       "      <td>Novice User</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>01986072-719a-404c-ae98-009d92e82323</td>\n",
       "      <td>2024-04-07</td>\n",
       "      <td>21.280727</td>\n",
       "      <td>10</td>\n",
       "      <td>15.372990</td>\n",
       "      <td>121</td>\n",
       "      <td>8.417014</td>\n",
       "      <td>0.519473</td>\n",
       "      <td>0.368913</td>\n",
       "      <td>2.108141</td>\n",
       "      <td>...</td>\n",
       "      <td>Level4</td>\n",
       "      <td>IT</td>\n",
       "      <td>Junior IC</td>\n",
       "      <td>12</td>\n",
       "      <td>11.00</td>\n",
       "      <td>11.00</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>Novice User</td>\n",
       "      <td>Novice User</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>01986072-719a-404c-ae98-009d92e82323</td>\n",
       "      <td>2024-04-14</td>\n",
       "      <td>17.450330</td>\n",
       "      <td>8</td>\n",
       "      <td>11.808617</td>\n",
       "      <td>104</td>\n",
       "      <td>7.889519</td>\n",
       "      <td>1.907069</td>\n",
       "      <td>0.096829</td>\n",
       "      <td>0.853150</td>\n",
       "      <td>...</td>\n",
       "      <td>Level4</td>\n",
       "      <td>IT</td>\n",
       "      <td>Junior IC</td>\n",
       "      <td>11</td>\n",
       "      <td>11.00</td>\n",
       "      <td>11.00</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>Novice User</td>\n",
       "      <td>Novice User</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>01986072-719a-404c-ae98-009d92e82323</td>\n",
       "      <td>2024-04-21</td>\n",
       "      <td>21.368059</td>\n",
       "      <td>3</td>\n",
       "      <td>14.908550</td>\n",
       "      <td>115</td>\n",
       "      <td>6.776404</td>\n",
       "      <td>0.209775</td>\n",
       "      <td>3.953832</td>\n",
       "      <td>0.878616</td>\n",
       "      <td>...</td>\n",
       "      <td>Level4</td>\n",
       "      <td>IT</td>\n",
       "      <td>Junior IC</td>\n",
       "      <td>4</td>\n",
       "      <td>9.25</td>\n",
       "      <td>9.25</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>Novice User</td>\n",
       "      <td>Habitual User</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>01986072-719a-404c-ae98-009d92e82323</td>\n",
       "      <td>2024-04-28</td>\n",
       "      <td>20.849744</td>\n",
       "      <td>5</td>\n",
       "      <td>13.737000</td>\n",
       "      <td>110</td>\n",
       "      <td>8.759793</td>\n",
       "      <td>0.931585</td>\n",
       "      <td>1.201305</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>Level4</td>\n",
       "      <td>IT</td>\n",
       "      <td>Junior IC</td>\n",
       "      <td>6</td>\n",
       "      <td>8.60</td>\n",
       "      <td>8.25</td>\n",
       "      <td>False</td>\n",
       "      <td>True</td>\n",
       "      <td>Novice User</td>\n",
       "      <td>Habitual User</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 80 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                               PersonId MetricDate  Collaboration_hours  \\\n",
       "0  01986072-719a-404c-ae98-009d92e82323 2024-03-31            26.884733   \n",
       "1  01986072-719a-404c-ae98-009d92e82323 2024-04-07            21.280727   \n",
       "2  01986072-719a-404c-ae98-009d92e82323 2024-04-14            17.450330   \n",
       "3  01986072-719a-404c-ae98-009d92e82323 2024-04-21            21.368059   \n",
       "4  01986072-719a-404c-ae98-009d92e82323 2024-04-28            20.849744   \n",
       "\n",
       "   Copilot_actions_taken_in_Teams  Meeting_and_call_hours  \\\n",
       "0                               7               17.700027   \n",
       "1                              10               15.372990   \n",
       "2                               8               11.808617   \n",
       "3                               3               14.908550   \n",
       "4                               5               13.737000   \n",
       "\n",
       "   Internal_network_size  Email_hours  Channel_message_posts  \\\n",
       "0                    156     9.667004               0.117751   \n",
       "1                    121     8.417014               0.519473   \n",
       "2                    104     7.889519               1.907069   \n",
       "3                    115     6.776404               0.209775   \n",
       "4                    110     8.759793               0.931585   \n",
       "\n",
       "   Conflicting_meeting_hours  Large_and_long_meeting_hours  ...   Level  \\\n",
       "0                   2.674868                      1.262361  ...  Level4   \n",
       "1                   0.368913                      2.108141  ...  Level4   \n",
       "2                   0.096829                      0.853150  ...  Level4   \n",
       "3                   3.953832                      0.878616  ...  Level4   \n",
       "4                   1.201305                      0.000000  ...  Level4   \n",
       "\n",
       "   Organization  LevelDesignation  target_metric  target_metric_l12w  \\\n",
       "0            IT         Junior IC             10               10.00   \n",
       "1            IT         Junior IC             12               11.00   \n",
       "2            IT         Junior IC             11               11.00   \n",
       "3            IT         Junior IC              4                9.25   \n",
       "4            IT         Junior IC              6                8.60   \n",
       "\n",
       "   target_metric_l4w  IsHabit12w  IsHabit4w  UsageSegments_12w  \\\n",
       "0              10.00       False      False        Novice User   \n",
       "1              11.00       False      False        Novice User   \n",
       "2              11.00       False      False        Novice User   \n",
       "3               9.25       False       True        Novice User   \n",
       "4               8.25       False       True        Novice User   \n",
       "\n",
       "   UsageSegments_4w  \n",
       "0       Novice User  \n",
       "1       Novice User  \n",
       "2       Novice User  \n",
       "3     Habitual User  \n",
       "4     Habitual User  \n",
       "\n",
       "[5 rows x 80 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Identify usage segments\n",
    "usage_segments_data = vi.identify_usage_segments(\n",
    "    data=pq_data,\n",
    "    metric_str=[\n",
    "        \"Copilot_actions_taken_in_Teams\",\n",
    "        \"Copilot_actions_taken_in_Outlook\",\n",
    "        \"Copilot_actions_taken_in_Excel\",\n",
    "        \"Copilot_actions_taken_in_Word\",\n",
    "        \"Copilot_actions_taken_in_Powerpoint\"\n",
    "    ],\n",
    "    version=\"12w\",\n",
    "    return_type=\"data\"\n",
    ")\n",
    "\n",
    "# Display the first few rows of the updated dataset\n",
    "usage_segments_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e38a9ffe",
   "metadata": {},
   "source": [
    "### Visualize the mean of `target_metric` by usage segment\n",
    "\n",
    "To better understand usage behavior, create a bar plot showing the mean of `target_metric` grouped by `UsageSegments_12w`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "df3b0152",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 400x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize the mean of `target_metric` by `UsageSegments_12w`\n",
    "usage_segments_bar_plot = vi.create_bar(\n",
    "    data=usage_segments_data,\n",
    "    metric=\"target_metric\",\n",
    "    hrvar=\"UsageSegments_12w\",\n",
    "    return_type=\"plot\",\n",
    "    plot_title=\"Mean Target Metric by Usage Segment\",\n",
    "    plot_subtitle=\"Based on 12-week rolling averages\"\n",
    ")\n",
    "\n",
    "# Display the bar plot\n",
    "usage_segments_bar_plot.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "07b9c4e2",
   "metadata": {},
   "source": [
    "### Visualize usage segments over time\n",
    "\n",
    "Next, visualize the distribution of usage segments over time using `identify_usage_segments` with `return_type='plot'`. The following shows a horizontal stacked bar plot, which shows the evolution in the proportion of the usage segments over time. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "a5efff68",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize usage segments over time\n",
    "usage_segments_time_plot = vi.identify_usage_segments(\n",
    "    data=pq_data,\n",
    "    metric_str=[\n",
    "        \"Copilot_actions_taken_in_Teams\",\n",
    "        \"Copilot_actions_taken_in_Outlook\",\n",
    "        \"Copilot_actions_taken_in_Excel\",\n",
    "        \"Copilot_actions_taken_in_Word\",\n",
    "        \"Copilot_actions_taken_in_Powerpoint\"\n",
    "    ],\n",
    "    version=\"12w\",\n",
    "    return_type=\"plot\"\n",
    ")\n",
    "\n",
    "# Display the time plot\n",
    "usage_segments_time_plot.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "18574916",
   "metadata": {},
   "source": [
    "## Step 3: Compute favorability scores for ordinal metrics\n",
    "\n",
    "Before calculating odds ratios, use `compute_fav()` to convert ordinal metrics into categorical variables representing favorable and unfavorable scores. This standardizes metrics to a 100-point scale, making results easier to interpret and compare.\n",
    "\n",
    "Neutral scores are dropped to focus on the most meaningful responses.\n",
    "\n",
    "In `usage_segments_data` printed below, it can be seen that `compute_fav()` has added several columns suffixing the `ordinal_metrics` columns with `_100` and `_fav`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "2b225ab2",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>PersonId</th>\n",
       "      <th>MetricDate</th>\n",
       "      <th>Collaboration_hours</th>\n",
       "      <th>Copilot_actions_taken_in_Teams</th>\n",
       "      <th>Meeting_and_call_hours</th>\n",
       "      <th>Internal_network_size</th>\n",
       "      <th>Email_hours</th>\n",
       "      <th>Channel_message_posts</th>\n",
       "      <th>Conflicting_meeting_hours</th>\n",
       "      <th>Large_and_long_meeting_hours</th>\n",
       "      <th>...</th>\n",
       "      <th>Resources_100</th>\n",
       "      <th>Resources_fav</th>\n",
       "      <th>Speak_My_Mind_100</th>\n",
       "      <th>Speak_My_Mind_fav</th>\n",
       "      <th>Wellbeing_100</th>\n",
       "      <th>Wellbeing_fav</th>\n",
       "      <th>Work_Life_Balance_100</th>\n",
       "      <th>Work_Life_Balance_fav</th>\n",
       "      <th>Workload_100</th>\n",
       "      <th>Workload_fav</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>02723512-4f45-4385-8d1a-c23048e1e961</td>\n",
       "      <td>2024-04-07</td>\n",
       "      <td>26.310260</td>\n",
       "      <td>1</td>\n",
       "      <td>17.635230</td>\n",
       "      <td>124</td>\n",
       "      <td>10.887553</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.322255</td>\n",
       "      <td>0.067661</td>\n",
       "      <td>...</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>100.0</td>\n",
       "      <td>fav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>02c55079-f137-4abb-9806-f58e9b60efd6</td>\n",
       "      <td>2024-06-30</td>\n",
       "      <td>17.401642</td>\n",
       "      <td>4</td>\n",
       "      <td>10.399207</td>\n",
       "      <td>84</td>\n",
       "      <td>5.253439</td>\n",
       "      <td>0.195852</td>\n",
       "      <td>3.203440</td>\n",
       "      <td>0.975272</td>\n",
       "      <td>...</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>100.0</td>\n",
       "      <td>fav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>123</th>\n",
       "      <td>02ddc980-8f37-4156-9397-6d621e445a00</td>\n",
       "      <td>2024-08-04</td>\n",
       "      <td>20.612899</td>\n",
       "      <td>3</td>\n",
       "      <td>14.130869</td>\n",
       "      <td>103</td>\n",
       "      <td>8.070390</td>\n",
       "      <td>0.577123</td>\n",
       "      <td>1.374351</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>100.0</td>\n",
       "      <td>fav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>135</th>\n",
       "      <td>02ddc980-8f37-4156-9397-6d621e445a00</td>\n",
       "      <td>2024-10-27</td>\n",
       "      <td>19.514361</td>\n",
       "      <td>2</td>\n",
       "      <td>10.986860</td>\n",
       "      <td>91</td>\n",
       "      <td>6.221707</td>\n",
       "      <td>2.286118</td>\n",
       "      <td>2.294472</td>\n",
       "      <td>0.391576</td>\n",
       "      <td>...</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>100.0</td>\n",
       "      <td>fav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>164</th>\n",
       "      <td>032432ad-390c-4ce4-9f25-d5be080bd982</td>\n",
       "      <td>2024-09-15</td>\n",
       "      <td>34.160594</td>\n",
       "      <td>3</td>\n",
       "      <td>27.364673</td>\n",
       "      <td>182</td>\n",
       "      <td>12.926987</td>\n",
       "      <td>0.197464</td>\n",
       "      <td>6.306590</td>\n",
       "      <td>1.153810</td>\n",
       "      <td>...</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>25.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>100.0</td>\n",
       "      <td>fav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "      <td>0.0</td>\n",
       "      <td>unfav</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 96 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 PersonId MetricDate  Collaboration_hours  \\\n",
       "36   02723512-4f45-4385-8d1a-c23048e1e961 2024-04-07            26.310260   \n",
       "83   02c55079-f137-4abb-9806-f58e9b60efd6 2024-06-30            17.401642   \n",
       "123  02ddc980-8f37-4156-9397-6d621e445a00 2024-08-04            20.612899   \n",
       "135  02ddc980-8f37-4156-9397-6d621e445a00 2024-10-27            19.514361   \n",
       "164  032432ad-390c-4ce4-9f25-d5be080bd982 2024-09-15            34.160594   \n",
       "\n",
       "     Copilot_actions_taken_in_Teams  Meeting_and_call_hours  \\\n",
       "36                                1               17.635230   \n",
       "83                                4               10.399207   \n",
       "123                               3               14.130869   \n",
       "135                               2               10.986860   \n",
       "164                               3               27.364673   \n",
       "\n",
       "     Internal_network_size  Email_hours  Channel_message_posts  \\\n",
       "36                     124    10.887553               0.000000   \n",
       "83                      84     5.253439               0.195852   \n",
       "123                    103     8.070390               0.577123   \n",
       "135                     91     6.221707               2.286118   \n",
       "164                    182    12.926987               0.197464   \n",
       "\n",
       "     Conflicting_meeting_hours  Large_and_long_meeting_hours  ...  \\\n",
       "36                    3.322255                      0.067661  ...   \n",
       "83                    3.203440                      0.975272  ...   \n",
       "123                   1.374351                      0.000000  ...   \n",
       "135                   2.294472                      0.391576  ...   \n",
       "164                   6.306590                      1.153810  ...   \n",
       "\n",
       "     Resources_100  Resources_fav  Speak_My_Mind_100  Speak_My_Mind_fav  \\\n",
       "36            25.0          unfav               25.0              unfav   \n",
       "83            25.0          unfav               25.0              unfav   \n",
       "123           25.0          unfav               25.0              unfav   \n",
       "135           25.0          unfav               25.0              unfav   \n",
       "164           25.0          unfav               25.0              unfav   \n",
       "\n",
       "     Wellbeing_100  Wellbeing_fav  Work_Life_Balance_100  \\\n",
       "36           100.0            fav                    0.0   \n",
       "83           100.0            fav                    0.0   \n",
       "123          100.0            fav                    0.0   \n",
       "135          100.0            fav                    0.0   \n",
       "164          100.0            fav                    0.0   \n",
       "\n",
       "     Work_Life_Balance_fav  Workload_100  Workload_fav  \n",
       "36                   unfav           0.0         unfav  \n",
       "83                   unfav           0.0         unfav  \n",
       "123                  unfav           0.0         unfav  \n",
       "135                  unfav           0.0         unfav  \n",
       "164                  unfav           0.0         unfav  \n",
       "\n",
       "[5 rows x 96 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Define the ordinal metrics\n",
    "ordinal_metrics = [\n",
    "    \"eSat\",\n",
    "    \"Initiative\",\n",
    "    \"Manager_Recommend\",\n",
    "    \"Resources\",\n",
    "    \"Speak_My_Mind\",\n",
    "    \"Wellbeing\",\n",
    "    \"Work_Life_Balance\",\n",
    "    \"Workload\"\n",
    "]\n",
    "\n",
    "# Compute favorability scores\n",
    "usage_segments_data = vi.compute_fav(\n",
    "    data=usage_segments_data,\n",
    "    ord_metrics=ordinal_metrics,\n",
    "    item_options=5,  # Assuming a 5-point scale for ordinal metrics\n",
    "    fav_threshold=70,\n",
    "    unfav_threshold=40,\n",
    "    drop_neutral=True\n",
    ")\n",
    "\n",
    "# Display the first few rows of the updated dataset\n",
    "usage_segments_data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "609e59a4",
   "metadata": {},
   "source": [
    "## Step 4: Calculate odds ratios for ordinal metrics\n",
    "\n",
    "Now, calculate odds ratios for the favorability scores of these ordinal metrics:\n",
    "- `eSat`\n",
    "- `Initiative`\n",
    "- `Manager_Recommend`\n",
    "- `Resources`\n",
    "- `Speak_My_Mind`\n",
    "- `Wellbeing`\n",
    "- `Work_Life_Balance`\n",
    "- `Workload`\n",
    "\n",
    "The independent variable is `UsageSegments_12w`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "b5beb57a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   UsageSegments_12w Level  Odds_Ratio     Ordinal_Metric      n\n",
      "0      Habitual User     1    1.000000               eSat    3.0\n",
      "1        Novice User     1    1.000000               eSat    1.0\n",
      "2      Habitual User     2   53.571429               eSat  135.0\n",
      "3        Novice User     2   37.000000               eSat   49.0\n",
      "4      Habitual User     4   18.142857               eSat   58.0\n",
      "5        Novice User     4   10.333333               eSat   14.0\n",
      "6      Habitual User     5    0.428571               eSat    1.0\n",
      "7        Novice User     5    0.333333               eSat    NaN\n",
      "8      Habitual User     1    1.000000         Initiative    4.0\n",
      "9        Novice User     1    1.000000         Initiative    3.0\n",
      "10     Habitual User     2   54.333333         Initiative  166.0\n",
      "11       Novice User     2   18.142857         Initiative   57.0\n",
      "12     Habitual User     4    1.444444         Initiative    6.0\n",
      "13       Novice User     4    1.571429         Initiative    5.0\n",
      "14     Habitual User     1    1.000000  Manager_Recommend    5.0\n",
      "15       Novice User     1    1.000000  Manager_Recommend    NaN\n",
      "16     Habitual User     2   43.545455  Manager_Recommend  163.0\n",
      "17       Novice User     2  133.000000  Manager_Recommend   59.0\n",
      "18     Habitual User     4    1.909091  Manager_Recommend   10.0\n",
      "19       Novice User     4    7.000000  Manager_Recommend    3.0\n",
      "20     Habitual User     5    0.090909  Manager_Recommend    NaN\n",
      "21       Novice User     5    5.000000  Manager_Recommend    2.0\n",
      "22     Habitual User     1    1.000000          Resources    2.0\n",
      "23       Novice User     1    1.000000          Resources    NaN\n",
      "24     Habitual User     2  100.600000          Resources  171.0\n",
      "25       Novice User     2  141.000000          Resources   62.0\n",
      "26     Habitual User     4    0.600000          Resources    1.0\n",
      "27       Novice User     4    3.000000          Resources    1.0\n",
      "28     Habitual User     1    1.000000      Speak_My_Mind    2.0\n",
      "29       Novice User     1    1.000000      Speak_My_Mind    1.0\n",
      "30     Habitual User     2   97.400000      Speak_My_Mind  166.0\n",
      "31       Novice User     2   43.666667      Speak_My_Mind   58.0\n",
      "32     Habitual User     4    3.800000      Speak_My_Mind    9.0\n",
      "33       Novice User     4    3.666667      Speak_My_Mind    5.0\n",
      "34     Habitual User     4    1.000000          Wellbeing   74.0\n",
      "35       Novice User     4    1.000000          Wellbeing   24.0\n",
      "36     Habitual User     5    1.786885          Wellbeing  123.0\n",
      "37       Novice User     5    1.938776          Wellbeing   43.0\n",
      "38     Habitual User     1    1.000000  Work_Life_Balance  143.0\n",
      "39       Novice User     1    1.000000  Work_Life_Balance   49.0\n",
      "40     Habitual User     2    0.352785  Work_Life_Balance   58.0\n",
      "41       Novice User     2    0.321101  Work_Life_Balance   17.0\n",
      "42     Habitual User     1    1.000000           Workload  143.0\n",
      "43       Novice User     1    1.000000           Workload   51.0\n",
      "44     Habitual User     2    0.317829           Workload   54.0\n",
      "45       Novice User     2    0.252174           Workload   14.0\n"
     ]
    }
   ],
   "source": [
    "# Calculate odds ratios\n",
    "odds_ratios_table = vi.create_odds_ratios(\n",
    "    data=usage_segments_data,\n",
    "    ord_metrics=ordinal_metrics,\n",
    "    metric=\"UsageSegments_12w\",\n",
    "    return_type=\"table\"\n",
    ")\n",
    "\n",
    "# Display the odds ratios table\n",
    "print(odds_ratios_table)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "03a56d54",
   "metadata": {},
   "source": [
    "Since favorability columns with the values `fav`, `unfav`, and `neu` have already been created using `compute_fav()`, you can use these directly in the proportional odds model to simplify the analysis.\n",
    "\n",
    "When interpreting odds ratios, a value greater than 1 indicates that the odds of a favorable outcome are higher for the group compared to the reference group, while a value less than 1 means the odds are lower. An odds ratio of exactly 1 suggests no difference between groups. This helps you understand how different usage segments are associated with the likelihood of favorable responses on each metric."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "98f2b746",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   UsageSegments_12w  Level  Odds_Ratio         Ordinal_Metric    n\n",
      "0      Habitual User    fav    1.000000               eSat_fav   59\n",
      "1        Novice User    fav    1.000000               eSat_fav   14\n",
      "2      Habitual User  unfav    2.953488               eSat_fav  137\n",
      "3        Novice User  unfav    3.645161               eSat_fav   50\n",
      "4      Habitual User    fav    1.000000         Initiative_fav    6\n",
      "5        Novice User    fav    1.000000         Initiative_fav    5\n",
      "6      Habitual User  unfav   38.230769         Initiative_fav  168\n",
      "7        Novice User  unfav   12.090909         Initiative_fav   59\n",
      "8      Habitual User    fav    1.000000  Manager_Recommend_fav   10\n",
      "9        Novice User    fav    1.000000  Manager_Recommend_fav    5\n",
      "10     Habitual User  unfav   23.285714  Manager_Recommend_fav  167\n",
      "11       Novice User  unfav   12.090909  Manager_Recommend_fav   59\n",
      "12     Habitual User    fav    1.000000          Resources_fav    1\n",
      "13       Novice User    fav    1.000000          Resources_fav    1\n",
      "14     Habitual User  unfav  169.000000          Resources_fav  172\n",
      "15       Novice User  unfav   47.000000          Resources_fav   62\n",
      "16     Habitual User    fav    1.000000      Speak_My_Mind_fav    9\n",
      "17       Novice User    fav    1.000000      Speak_My_Mind_fav    5\n",
      "18     Habitual User  unfav   25.842105      Speak_My_Mind_fav  167\n",
      "19       Novice User  unfav   12.090909      Speak_My_Mind_fav   59\n",
      "20     Habitual User    fav    1.000000          Wellbeing_fav  172\n",
      "21       Novice User    fav    1.000000          Wellbeing_fav   63\n",
      "22     Habitual User  unfav    1.000000  Work_Life_Balance_fav  172\n",
      "23       Novice User  unfav    1.000000  Work_Life_Balance_fav   63\n",
      "24     Habitual User  unfav    1.000000           Workload_fav  172\n",
      "25       Novice User  unfav    1.000000           Workload_fav   63\n"
     ]
    }
   ],
   "source": [
    "# Define ordinal metrics with '_fav' suffix\n",
    "ordinal_metrics_fav = [f\"{metric}_fav\" for metric in ordinal_metrics]\n",
    "\n",
    "# Calculate odds ratios\n",
    "odds_ratios_table_fav = vi.create_odds_ratios(\n",
    "    data=usage_segments_data,\n",
    "    ord_metrics=ordinal_metrics_fav,\n",
    "    metric=\"UsageSegments_12w\",\n",
    "    return_type=\"table\"\n",
    ")\n",
    "\n",
    "# Display the odds ratios table\n",
    "print(odds_ratios_table_fav)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "ffad1e4a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   UsageSegments_12w Level  Odds_Ratio         Ordinal_Metric    n\n",
      "4      Habitual User   fav         1.0         Initiative_fav    6\n",
      "5        Novice User   fav         1.0         Initiative_fav    5\n",
      "8      Habitual User   fav         1.0  Manager_Recommend_fav   10\n",
      "16     Habitual User   fav         1.0      Speak_My_Mind_fav    9\n",
      "9        Novice User   fav         1.0  Manager_Recommend_fav    5\n",
      "12     Habitual User   fav         1.0          Resources_fav    1\n",
      "1        Novice User   fav         1.0               eSat_fav   14\n",
      "17       Novice User   fav         1.0      Speak_My_Mind_fav    5\n",
      "13       Novice User   fav         1.0          Resources_fav    1\n",
      "21       Novice User   fav         1.0          Wellbeing_fav   63\n",
      "0      Habitual User   fav         1.0               eSat_fav   59\n",
      "20     Habitual User   fav         1.0          Wellbeing_fav  172\n"
     ]
    }
   ],
   "source": [
    "# Filter for Level == 'fav' only, and sort Odds_Ratio in descending order\n",
    "odds_ratios_table_fav = odds_ratios_table_fav[odds_ratios_table_fav['Level'] == 'fav']\n",
    "odds_ratios_table_fav = odds_ratios_table_fav.sort_values(by='Odds_Ratio', ascending=False)\n",
    "\n",
    "print(odds_ratios_table_fav)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ba3e8ddf",
   "metadata": {},
   "source": [
    "## Step 5: Visualize the odds ratios\n",
    "\n",
    "Create a bar plot to visualize the odds ratios for the ordinal metrics, making it easier to compare the impact of usage segments."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "741925a0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize odds ratios\n",
    "odds_ratios_plot = vi.create_odds_ratios(\n",
    "    data=usage_segments_data,\n",
    "    ord_metrics=ordinal_metrics,\n",
    "    metric=\"UsageSegments_12w\",\n",
    "    return_type=\"plot\"\n",
    ")\n",
    "\n",
    "# Display the plot\n",
    "odds_ratios_plot.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "b5f73adb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Visualize odds ratios for favorability\n",
    "odds_ratios_plot_fav = vi.create_odds_ratios(\n",
    "    data=usage_segments_data,\n",
    "    ord_metrics=ordinal_metrics_fav,\n",
    "    metric=\"UsageSegments_12w\",\n",
    "    return_type=\"plot\"\n",
    ")\n",
    "\n",
    "# Display the plot\n",
    "odds_ratios_plot_fav.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0b9da758",
   "metadata": {},
   "source": [
    "## Summary\n",
    "\n",
    "In this notebook, you learned how to:\n",
    "1. Load demo data (`pq_data`).\n",
    "2. Create an independent variable (`UsageSegments_12w`) using `identify_usage_segments`.\n",
    "3. Compute favorability scores for ordinal metrics with `compute_fav`.\n",
    "4. Calculate odds ratios for ordinal metrics using `create_odds_ratios`.\n",
    "5. Visualize the results for interpretation.\n",
    "\n",
    "By combining `create_odds_ratios` with `compute_fav`, you can consistently analyze the relationship between ordinal metrics and independent variables, regardless of the original point scale."
   ]
  }
 ],
 "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": "3.12.10"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}