{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6dbd4ad9",
   "metadata": {},
   "source": [
    "# Secondary RQ — Regular public transit & communication apprehension\n",
    "\n",
    "**Sample:** Prolific File A + File B stacked ∩ Qualtrics File C, matched on `Q0` / `Participant id`, with complete PRCA group + interpersonal ground truth.\n",
    "\n",
    "**Secondary research question:** Do individuals who take public transportation **regularly** have communication-apprehension (CA) scores that differ statistically from the larger matched cohort? If so, by how much — and what does the score distribution look like?\n",
    "\n",
    "**Primary exposure (Q26):** public transportation days in the last 3 months.  \n",
    "**Regular riders** = `4-8 days a month` **or** `8 or more days a month` (weekly-or-more).\n",
    "\n",
    "This notebook:\n",
    "\n",
    "1. Loads / cleans the matched analytic sample  \n",
    "2. Labels regular vs non-regular riders  \n",
    "3. Describes CA distributions  \n",
    "4. Runs Welch *t*-tests + Mann–Whitney + bootstrap CIs + effect sizes  \n",
    "5. Checks sensitivity across alternate Q26 cutoffs  \n",
    "6. Writes distributable artifacts under `outputs/transit_ca/`\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "a902a547",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:30.939529Z",
     "iopub.status.busy": "2026-07-25T15:42:30.939405Z",
     "iopub.status.idle": "2026-07-25T15:42:31.689968Z",
     "shell.execute_reply": "2026-07-25T15:42:31.689170Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regular-rider labels: ['4-8 days a month', '8 or more days a month']\n"
     ]
    }
   ],
   "source": [
    "from __future__ import annotations\n",
    "\n",
    "import json\n",
    "import sys\n",
    "from pathlib import Path\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "\n",
    "ROOT = Path.cwd()\n",
    "if ROOT.name == \"notebooks\":\n",
    "    ROOT = ROOT.parent\n",
    "SRC = ROOT / \"src\"\n",
    "if str(SRC) not in sys.path:\n",
    "    sys.path.insert(0, str(SRC))\n",
    "\n",
    "from ca_personas.load import load_full_cohort\n",
    "from ca_personas.paths import sibling_data_available\n",
    "from ca_personas.transit_ca import (\n",
    "    PRIMARY_REGULAR_LABELS,\n",
    "    run_transit_ca_analysis,\n",
    "    save_transit_ca_artifacts,\n",
    ")\n",
    "\n",
    "pd.set_option(\"display.max_columns\", 50)\n",
    "pd.set_option(\"display.float_format\", lambda x: f\"{x:0.4f}\")\n",
    "plt.rcParams.update({\n",
    "    \"axes.spines.top\": False,\n",
    "    \"axes.spines.right\": False,\n",
    "    \"axes.grid\": True,\n",
    "    \"grid.alpha\": 0.25,\n",
    "})\n",
    "\n",
    "assert sibling_data_available(), (\n",
    "    \"Place File A/B/C under ../sibling_data/ before running this notebook.\"\n",
    ")\n",
    "print(\"Regular-rider labels:\", sorted(PRIMARY_REGULAR_LABELS))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "f8ca6f7b",
   "metadata": {},
   "source": [
    "## 1. Load matched analytic sample\n",
    "\n",
    "Inner join of stacked Prolific waves with Qualtrics File C; keep respondents with complete PRCA items (`Q1–Q6`, `Q13–Q18`).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "7dddb2f7",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:31.692031Z",
     "iopub.status.busy": "2026-07-25T15:42:31.691864Z",
     "iopub.status.idle": "2026-07-25T15:42:32.009134Z",
     "shell.execute_reply": "2026-07-25T15:42:32.008499Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{\n",
      "  \"n_prolific_raw\": 262,\n",
      "  \"n_prolific_unique\": 262,\n",
      "  \"n_qualtrics_raw\": 273,\n",
      "  \"n_qualtrics_with_pid\": 255,\n",
      "  \"n_qualtrics_complete_ca\": 260,\n",
      "  \"n_joined\": 252,\n",
      "  \"n_matched_both\": 252,\n",
      "  \"n_analytic\": 241,\n",
      "  \"n_dropped_missing_pid\": 0,\n",
      "  \"n_dropped_incomplete_ca\": 11,\n",
      "  \"n_dropped_unscorable_ca\": 0,\n",
      "  \"n_dropped_unjoined\": 31,\n",
      "  \"n_prolific_only\": 10,\n",
      "  \"n_qualtrics_only\": 21,\n",
      "  \"n_qualtrics_missing_pid\": 18,\n",
      "  \"waves\": {\n",
      "    \"B\": 153,\n",
      "    \"A\": 99\n",
      "  },\n",
      "  \"notes\": [\n",
      "    \"Normalized 19 DATA_EXPIRED Student status values to missing.\",\n",
      "    \"Full Prolific waves (File A/B) omit Ethnicity / Nationality / Language; demos tier uses Age, Sex, Country of residence, and Student status.\",\n",
      "    \"Analytic sample = Prolific\\u2229Qualtrics with complete scorable PRCA group + interpersonal items.\",\n",
      "    \"Merge coverage (pre-CA filter): 252 matched Prolific\\u2229Qualtrics; 21 Qualtrics-only (incl. 18 blank Q0 test rows; disregard); 10 Prolific-only (disregard).\"\n",
      "  ]\n",
      "}\n",
      "Analytic N: 241\n"
     ]
    },
    {
     "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>participant_id</th>\n",
       "      <th>Q26</th>\n",
       "      <th>gt_group_ca</th>\n",
       "      <th>gt_interpersonal_ca</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>ebab12e9878344fdc93a8c1cad27ff2f709e12b90f0496...</td>\n",
       "      <td>2-4 days a month</td>\n",
       "      <td>14</td>\n",
       "      <td>16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>b6954799ca47bc836ac02ab86e4710c3b0e096d7ace449...</td>\n",
       "      <td>Never</td>\n",
       "      <td>17</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>6ac07c3845b0ce5cac4ab3569ea70236b7cfa8fd51576e...</td>\n",
       "      <td>Never</td>\n",
       "      <td>11</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>c67b3266ecd4af12979c1f67d1efffd0d3e9bfbc567d21...</td>\n",
       "      <td>Never</td>\n",
       "      <td>10</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1cefb8d34b92489e8ffd9cf34c469964f0527b1cae33e0...</td>\n",
       "      <td>Never</td>\n",
       "      <td>8</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                      participant_id               Q26  \\\n",
       "0  ebab12e9878344fdc93a8c1cad27ff2f709e12b90f0496...  2-4 days a month   \n",
       "1  b6954799ca47bc836ac02ab86e4710c3b0e096d7ace449...             Never   \n",
       "2  6ac07c3845b0ce5cac4ab3569ea70236b7cfa8fd51576e...             Never   \n",
       "3  c67b3266ecd4af12979c1f67d1efffd0d3e9bfbc567d21...             Never   \n",
       "4  1cefb8d34b92489e8ffd9cf34c469964f0527b1cae33e0...             Never   \n",
       "\n",
       "   gt_group_ca  gt_interpersonal_ca  \n",
       "0           14                   16  \n",
       "1           17                   17  \n",
       "2           11                   12  \n",
       "3           10                    6  \n",
       "4            8                   12  "
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "participants, cleaning_report = load_full_cohort(join_how=\"inner\")\n",
    "print(json.dumps(cleaning_report, indent=2))\n",
    "print(\"Analytic N:\", len(participants))\n",
    "participants[[\"participant_id\", \"Q26\", \"gt_group_ca\", \"gt_interpersonal_ca\"]].head()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e1f37904",
   "metadata": {},
   "source": [
    "## 2. Run secondary-RQ analysis pipeline\n",
    "\n",
    "Welch *t*-test (primary), Mann–Whitney (nonparametric sensitivity), Cohen's *d* / Hedges' *g*, and bootstrap 95% CIs for the mean difference (regular − not regular). Also reports regular mean vs overall cohort mean for the “how much vs the larger population” contrast.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "cb471810",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:32.011217Z",
     "iopub.status.busy": "2026-07-25T15:42:32.011078Z",
     "iopub.status.idle": "2026-07-25T15:42:32.484730Z",
     "shell.execute_reply": "2026-07-25T15:42:32.483778Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{\n",
      "  \"secondary_rq\": \"Do regular public-transit riders have CA scores that differ statistically from the larger matched cohort, and by how much?\",\n",
      "  \"sample\": {\n",
      "    \"n_analytic_input\": 241,\n",
      "    \"n_with_q26\": 241,\n",
      "    \"n_regular\": 101,\n",
      "    \"n_not_regular\": 140,\n",
      "    \"regular_definition\": \"Q26 in {4-8 days a month, 8 or more days a month} (weekly-or-more public transit)\",\n",
      "    \"regular_labels\": [\n",
      "      \"4-8 days a month\",\n",
      "      \"8 or more days a month\"\n",
      "    ]\n",
      "  },\n",
      "  \"primary_tests\": [\n",
      "    {\n",
      "      \"score\": \"gt_group_ca\",\n",
      "      \"n_regular\": 101,\n",
      "      \"n_not_regular\": 140,\n",
      "      \"n_overall\": 241,\n",
      "      \"mean_regular\": 13.03960396039604,\n",
      "      \"mean_not_regular\": 15.764285714285714,\n",
      "      \"mean_overall\": 14.622406639004149,\n",
      "      \"diff_regular_minus_not_regular\": -2.7246817538896746,\n",
      "      \"diff_regular_minus_overall\": -1.5828026786081093,\n",
      "      \"pct_diff_vs_overall\": -10.824501859947627,\n",
      "      \"welch_t\": -3.676531638712411,\n",
      "      \"welch_df\": 236.9131613823941,\n",
      "      \"welch_p\": 0.00029250822800505247,\n",
      "      \"mannwhitney_u\": 5374.5,\n",
      "      \"mannwhitney_p\": 0.0014555822517995248,\n",
      "      \"cohens_d\": -0.46238123740548615,\n",
      "      \"hedges_g\": -0.4609287309005474,\n",
      "      \"boot_diff_vs_not_regular\": -2.7246817538896746,\n",
      "      \"boot_ci_low\": -4.14541548797737,\n",
      "      \"boot_ci_high\": -1.2844236209335258,\n",
      "      \"significant_at_05\": true\n",
      "    },\n",
      "    {\n",
      "      \"score\": \"gt_interpersonal_ca\",\n",
      "      \"n_regular\": 101,\n",
      "      \"n_not_regular\": 140,\n",
      "      \"n_overall\": 241,\n",
      "      \"mean_regular\": 13.306930693069306,\n",
      "      \"mean_not_regular\": 15.035714285714286,\n",
      "      \"mean_overall\": 14.311203319502075,\n",
      "      \"diff_regular_minus_not_regular\": -1.7287835926449802,\n",
      "      \"diff_regular_minus_overall\": -1.004272626432769,\n",
      "      \"pct_diff_vs_overall\": -7.017387734714332,\n",
      "      \"welch_t\": -2.3405333419995706,\n",
      "      \"welch_df\": 228.7997232360521,\n",
      "      \"welch_p\": 0.020116568031051666,\n",
      "      \"mannwhitney_u\": 5875.0,\n",
      "      \"mannwhitney_p\": 0.024928892975388507,\n",
      "      \"cohens_d\": -0.2997797392856572,\n",
      "      \"hedges_g\": -0.2988380228271682,\n",
      "      \"boot_diff_vs_not_regular\": -1.7287835926449802,\n",
      "      \"boot_ci_low\": -3.1220526874115992,\n",
      "      \"boot_ci_high\": -0.2536386138613856,\n",
      "      \"significant_at_05\": true\n",
      "    }\n",
      "  ],\n",
      "  \"verdict\": {\n",
      "    \"any_subscale_significant_at_05\": true,\n",
      "    \"interpretation\": \"At least one PRCA subscale differs significantly (Welch t, \\u03b1=.05) between regular and non-regular riders.\"\n",
      "  }\n",
      "}\n"
     ]
    }
   ],
   "source": [
    "analysis = run_transit_ca_analysis(participants, n_boot=5000, random_state=42)\n",
    "summary = analysis[\"summary\"]\n",
    "print(json.dumps(summary, indent=2))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "04c6c999",
   "metadata": {},
   "source": [
    "## 3. Exposure frequencies (Q26)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "fd06a666",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:32.486365Z",
     "iopub.status.busy": "2026-07-25T15:42:32.486249Z",
     "iopub.status.idle": "2026-07-25T15:42:32.491087Z",
     "shell.execute_reply": "2026-07-25T15:42:32.490345Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Q26\n",
      "8 or more days a month    55\n",
      "2-4 days a month          53\n",
      "Never                     50\n",
      "4-8 days a month          46\n",
      "0-1 days a month          37\n",
      "Name: count, dtype: int64\n",
      "\n",
      "transit_group\n",
      "not_regular    140\n",
      "regular        101\n",
      "Name: count, dtype: int64\n",
      "Regular share: 41.9% (101/241)\n"
     ]
    }
   ],
   "source": [
    "labeled = analysis[\"labeled\"]\n",
    "exposure = labeled.dropna(subset=[\"regular_transit\"])\n",
    "print(exposure[\"Q26\"].value_counts(dropna=False))\n",
    "print()\n",
    "print(exposure[\"transit_group\"].value_counts())\n",
    "print(\n",
    "    f\"Regular share: {(exposure['regular_transit'] == True).mean():.1%} \"\n",
    "    f\"({int((exposure['regular_transit'] == True).sum())}/{len(exposure)})\"\n",
    ")\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "18688722",
   "metadata": {},
   "source": [
    "## 4. Descriptive CA by transit group\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "af4291aa",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:32.492207Z",
     "iopub.status.busy": "2026-07-25T15:42:32.492077Z",
     "iopub.status.idle": "2026-07-25T15:42:32.496964Z",
     "shell.execute_reply": "2026-07-25T15:42:32.496499Z"
    }
   },
   "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>score</th>\n",
       "      <th>group</th>\n",
       "      <th>n</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>median</th>\n",
       "      <th>min</th>\n",
       "      <th>max</th>\n",
       "      <th>q25</th>\n",
       "      <th>q75</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>gt_group_ca</td>\n",
       "      <td>overall</td>\n",
       "      <td>241</td>\n",
       "      <td>14.6224</td>\n",
       "      <td>6.0328</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>6.0000</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>10.0000</td>\n",
       "      <td>18.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gt_group_ca</td>\n",
       "      <td>regular</td>\n",
       "      <td>101</td>\n",
       "      <td>13.0396</td>\n",
       "      <td>5.0772</td>\n",
       "      <td>12.0000</td>\n",
       "      <td>6.0000</td>\n",
       "      <td>27.0000</td>\n",
       "      <td>10.0000</td>\n",
       "      <td>16.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>gt_group_ca</td>\n",
       "      <td>not_regular</td>\n",
       "      <td>140</td>\n",
       "      <td>15.7643</td>\n",
       "      <td>6.4156</td>\n",
       "      <td>14.0000</td>\n",
       "      <td>6.0000</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>11.0000</td>\n",
       "      <td>20.2500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>gt_interpersonal_ca</td>\n",
       "      <td>overall</td>\n",
       "      <td>241</td>\n",
       "      <td>14.3112</td>\n",
       "      <td>5.8180</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>6.0000</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>10.0000</td>\n",
       "      <td>18.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>gt_interpersonal_ca</td>\n",
       "      <td>regular</td>\n",
       "      <td>101</td>\n",
       "      <td>13.3069</td>\n",
       "      <td>5.3661</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>6.0000</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>10.0000</td>\n",
       "      <td>16.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>gt_interpersonal_ca</td>\n",
       "      <td>not_regular</td>\n",
       "      <td>140</td>\n",
       "      <td>15.0357</td>\n",
       "      <td>6.0387</td>\n",
       "      <td>14.0000</td>\n",
       "      <td>6.0000</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>11.0000</td>\n",
       "      <td>18.2500</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 score        group    n    mean    std  median    min  \\\n",
       "0          gt_group_ca      overall  241 14.6224 6.0328 13.0000 6.0000   \n",
       "1          gt_group_ca      regular  101 13.0396 5.0772 12.0000 6.0000   \n",
       "2          gt_group_ca  not_regular  140 15.7643 6.4156 14.0000 6.0000   \n",
       "3  gt_interpersonal_ca      overall  241 14.3112 5.8180 13.0000 6.0000   \n",
       "4  gt_interpersonal_ca      regular  101 13.3069 5.3661 13.0000 6.0000   \n",
       "5  gt_interpersonal_ca  not_regular  140 15.0357 6.0387 14.0000 6.0000   \n",
       "\n",
       "      max     q25     q75  \n",
       "0 30.0000 10.0000 18.0000  \n",
       "1 27.0000 10.0000 16.0000  \n",
       "2 30.0000 11.0000 20.2500  \n",
       "3 30.0000 10.0000 18.0000  \n",
       "4 30.0000 10.0000 16.0000  \n",
       "5 30.0000 11.0000 18.2500  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "analysis[\"descriptives\"]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "916a0f24",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:32.498683Z",
     "iopub.status.busy": "2026-07-25T15:42:32.498546Z",
     "iopub.status.idle": "2026-07-25T15:42:32.503482Z",
     "shell.execute_reply": "2026-07-25T15:42:32.502882Z"
    }
   },
   "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>Q26</th>\n",
       "      <th>n</th>\n",
       "      <th>mean_group_ca</th>\n",
       "      <th>std_group_ca</th>\n",
       "      <th>mean_interpersonal_ca</th>\n",
       "      <th>std_interpersonal_ca</th>\n",
       "      <th>median_group_ca</th>\n",
       "      <th>median_interpersonal_ca</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Never</td>\n",
       "      <td>50</td>\n",
       "      <td>17.3800</td>\n",
       "      <td>6.8836</td>\n",
       "      <td>16.3400</td>\n",
       "      <td>6.9388</td>\n",
       "      <td>16.0000</td>\n",
       "      <td>16.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0-1 days a month</td>\n",
       "      <td>37</td>\n",
       "      <td>15.3784</td>\n",
       "      <td>5.8279</td>\n",
       "      <td>15.1622</td>\n",
       "      <td>5.2362</td>\n",
       "      <td>14.0000</td>\n",
       "      <td>14.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2-4 days a month</td>\n",
       "      <td>53</td>\n",
       "      <td>14.5094</td>\n",
       "      <td>6.1351</td>\n",
       "      <td>13.7170</td>\n",
       "      <td>5.4504</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>12.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4-8 days a month</td>\n",
       "      <td>46</td>\n",
       "      <td>13.2826</td>\n",
       "      <td>4.8884</td>\n",
       "      <td>13.4783</td>\n",
       "      <td>5.1585</td>\n",
       "      <td>12.0000</td>\n",
       "      <td>13.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>8 or more days a month</td>\n",
       "      <td>55</td>\n",
       "      <td>12.8364</td>\n",
       "      <td>5.2661</td>\n",
       "      <td>13.1636</td>\n",
       "      <td>5.5769</td>\n",
       "      <td>12.0000</td>\n",
       "      <td>12.0000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                      Q26   n  mean_group_ca  std_group_ca  \\\n",
       "0                   Never  50        17.3800        6.8836   \n",
       "1        0-1 days a month  37        15.3784        5.8279   \n",
       "2        2-4 days a month  53        14.5094        6.1351   \n",
       "3        4-8 days a month  46        13.2826        4.8884   \n",
       "4  8 or more days a month  55        12.8364        5.2661   \n",
       "\n",
       "   mean_interpersonal_ca  std_interpersonal_ca  median_group_ca  \\\n",
       "0                16.3400                6.9388          16.0000   \n",
       "1                15.1622                5.2362          14.0000   \n",
       "2                13.7170                5.4504          13.0000   \n",
       "3                13.4783                5.1585          12.0000   \n",
       "4                13.1636                5.5769          12.0000   \n",
       "\n",
       "   median_interpersonal_ca  \n",
       "0                  16.0000  \n",
       "1                  14.0000  \n",
       "2                  12.0000  \n",
       "3                  13.0000  \n",
       "4                  12.0000  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "analysis[\"by_q26\"]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "713288f7",
   "metadata": {},
   "source": [
    "## 5. Inferential tests — regular vs not-regular (and vs overall mean)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "3e5695ef",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:32.504541Z",
     "iopub.status.busy": "2026-07-25T15:42:32.504421Z",
     "iopub.status.idle": "2026-07-25T15:42:32.510542Z",
     "shell.execute_reply": "2026-07-25T15:42:32.510026Z"
    }
   },
   "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>score</th>\n",
       "      <th>n_regular</th>\n",
       "      <th>n_not_regular</th>\n",
       "      <th>mean_regular</th>\n",
       "      <th>mean_not_regular</th>\n",
       "      <th>mean_overall</th>\n",
       "      <th>diff_regular_minus_not_regular</th>\n",
       "      <th>diff_regular_minus_overall</th>\n",
       "      <th>pct_diff_vs_overall</th>\n",
       "      <th>welch_t</th>\n",
       "      <th>welch_p</th>\n",
       "      <th>mannwhitney_p</th>\n",
       "      <th>cohens_d</th>\n",
       "      <th>hedges_g</th>\n",
       "      <th>boot_ci_low</th>\n",
       "      <th>boot_ci_high</th>\n",
       "      <th>significant_at_05</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>gt_group_ca</td>\n",
       "      <td>101</td>\n",
       "      <td>140</td>\n",
       "      <td>13.0396</td>\n",
       "      <td>15.7643</td>\n",
       "      <td>14.6224</td>\n",
       "      <td>-2.7247</td>\n",
       "      <td>-1.5828</td>\n",
       "      <td>-10.8245</td>\n",
       "      <td>-3.6765</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>-0.4624</td>\n",
       "      <td>-0.4609</td>\n",
       "      <td>-4.1454</td>\n",
       "      <td>-1.2844</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gt_interpersonal_ca</td>\n",
       "      <td>101</td>\n",
       "      <td>140</td>\n",
       "      <td>13.3069</td>\n",
       "      <td>15.0357</td>\n",
       "      <td>14.3112</td>\n",
       "      <td>-1.7288</td>\n",
       "      <td>-1.0043</td>\n",
       "      <td>-7.0174</td>\n",
       "      <td>-2.3405</td>\n",
       "      <td>0.0201</td>\n",
       "      <td>0.0249</td>\n",
       "      <td>-0.2998</td>\n",
       "      <td>-0.2988</td>\n",
       "      <td>-3.1221</td>\n",
       "      <td>-0.2536</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 score  n_regular  n_not_regular  mean_regular  \\\n",
       "0          gt_group_ca        101            140       13.0396   \n",
       "1  gt_interpersonal_ca        101            140       13.3069   \n",
       "\n",
       "   mean_not_regular  mean_overall  diff_regular_minus_not_regular  \\\n",
       "0           15.7643       14.6224                         -2.7247   \n",
       "1           15.0357       14.3112                         -1.7288   \n",
       "\n",
       "   diff_regular_minus_overall  pct_diff_vs_overall  welch_t  welch_p  \\\n",
       "0                     -1.5828             -10.8245  -3.6765   0.0003   \n",
       "1                     -1.0043              -7.0174  -2.3405   0.0201   \n",
       "\n",
       "   mannwhitney_p  cohens_d  hedges_g  boot_ci_low  boot_ci_high  \\\n",
       "0         0.0015   -0.4624   -0.4609      -4.1454       -1.2844   \n",
       "1         0.0249   -0.2998   -0.2988      -3.1221       -0.2536   \n",
       "\n",
       "   significant_at_05  \n",
       "0               True  \n",
       "1               True  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "comps = analysis[\"comparisons\"].copy()\n",
    "show_cols = [\n",
    "    \"score\", \"n_regular\", \"n_not_regular\",\n",
    "    \"mean_regular\", \"mean_not_regular\", \"mean_overall\",\n",
    "    \"diff_regular_minus_not_regular\", \"diff_regular_minus_overall\",\n",
    "    \"pct_diff_vs_overall\",\n",
    "    \"welch_t\", \"welch_p\", \"mannwhitney_p\",\n",
    "    \"cohens_d\", \"hedges_g\",\n",
    "    \"boot_ci_low\", \"boot_ci_high\", \"significant_at_05\",\n",
    "]\n",
    "comps[show_cols]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f7afe5d5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:32.511550Z",
     "iopub.status.busy": "2026-07-25T15:42:32.511444Z",
     "iopub.status.idle": "2026-07-25T15:42:32.514721Z",
     "shell.execute_reply": "2026-07-25T15:42:32.514058Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Group CA: regular M=13.04 vs overall M=14.62 (Δ=-1.58 pts, -10.8%); vs non-regular Δ=-2.72 [95% boot CI -4.15, -1.28]; Welch p=0.0003; Cohen d=-0.462; significant@0.05=YES\n",
      "Interpersonal CA: regular M=13.31 vs overall M=14.31 (Δ=-1.00 pts, -7.0%); vs non-regular Δ=-1.73 [95% boot CI -3.12, -0.25]; Welch p=0.0201; Cohen d=-0.300; significant@0.05=YES\n",
      "\n",
      "Verdict: At least one PRCA subscale differs significantly (Welch t, α=.05) between regular and non-regular riders.\n"
     ]
    }
   ],
   "source": [
    "# Plain-language results card\n",
    "for row in summary[\"primary_tests\"]:\n",
    "    score = \"Group CA\" if row[\"score\"] == \"gt_group_ca\" else \"Interpersonal CA\"\n",
    "    direction = \"lower\" if row[\"diff_regular_minus_overall\"] < 0 else \"higher\"\n",
    "    sig = \"YES\" if row[\"significant_at_05\"] else \"no\"\n",
    "    print(\n",
    "        f\"{score}: regular M={row['mean_regular']:.2f} vs overall M={row['mean_overall']:.2f} \"\n",
    "        f\"(Δ={row['diff_regular_minus_overall']:+.2f} pts, {row['pct_diff_vs_overall']:+.1f}%); \"\n",
    "        f\"vs non-regular Δ={row['diff_regular_minus_not_regular']:+.2f} \"\n",
    "        f\"[95% boot CI {row['boot_ci_low']:+.2f}, {row['boot_ci_high']:+.2f}]; \"\n",
    "        f\"Welch p={row['welch_p']:.4f}; Cohen d={row['cohens_d']:.3f}; significant@0.05={sig}\"\n",
    "    )\n",
    "print()\n",
    "print(\"Verdict:\", summary[\"verdict\"][\"interpretation\"])\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ea0e0166",
   "metadata": {},
   "source": [
    "## 6. Distributions for reporting\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "98b97fb5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:32.515942Z",
     "iopub.status.busy": "2026-07-25T15:42:32.515839Z",
     "iopub.status.idle": "2026-07-25T15:42:32.838675Z",
     "shell.execute_reply": "2026-07-25T15:42:32.837723Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x450 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, 2, figsize=(11, 4.5), sharey=True)\n",
    "colors = {\"regular\": \"#C5050C\", \"not_regular\": \"#555555\"}\n",
    "for ax, score, title in zip(\n",
    "    axes,\n",
    "    [\"gt_group_ca\", \"gt_interpersonal_ca\"],\n",
    "    [\"Group CA\", \"Interpersonal CA\"],\n",
    "):\n",
    "    for group, color in colors.items():\n",
    "        vals = labeled.loc[labeled[\"transit_group\"] == group, score].dropna()\n",
    "        ax.hist(vals, bins=range(6, 32), alpha=0.55, label=group, color=color, edgecolor=\"white\")\n",
    "        ax.axvline(vals.mean(), color=color, linestyle=\"--\", linewidth=1.5)\n",
    "    ax.set_title(title)\n",
    "    ax.set_xlabel(\"PRCA subscale (6–30)\")\n",
    "axes[0].set_ylabel(\"Participants\")\n",
    "axes[0].legend(frameon=False, title=\"Transit group\")\n",
    "fig.suptitle(\"CA score distributions: regular vs not-regular public-transit riders\")\n",
    "fig.tight_layout()\n",
    "OUT = ROOT / \"outputs\" / \"transit_ca\"\n",
    "OUT.mkdir(parents=True, exist_ok=True)\n",
    "fig.savefig(OUT / \"fig_ca_distributions_by_transit.png\", dpi=150)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "fe642175",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:32.840250Z",
     "iopub.status.busy": "2026-07-25T15:42:32.840121Z",
     "iopub.status.idle": "2026-07-25T15:42:33.015384Z",
     "shell.execute_reply": "2026-07-25T15:42:33.014372Z"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1100x450 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig, axes = plt.subplots(1, 2, figsize=(11, 4.5), sharey=True)\n",
    "for ax, score, title in zip(\n",
    "    axes,\n",
    "    [\"gt_group_ca\", \"gt_interpersonal_ca\"],\n",
    "    [\"Group CA\", \"Interpersonal CA\"],\n",
    "):\n",
    "    data = [\n",
    "        labeled.loc[labeled[\"transit_group\"] == \"regular\", score].dropna(),\n",
    "        labeled.loc[labeled[\"transit_group\"] == \"not_regular\", score].dropna(),\n",
    "        labeled[score].dropna(),\n",
    "    ]\n",
    "    bp = ax.boxplot(data, tick_labels=[\"Regular\", \"Not regular\", \"Overall\"], patch_artist=True)\n",
    "    for patch, color in zip(bp[\"boxes\"], [\"#C5050C\", \"#888888\", \"#1a1a1a\"]):\n",
    "        patch.set_facecolor(color)\n",
    "        patch.set_alpha(0.35)\n",
    "    ax.set_title(title)\n",
    "    ax.set_ylabel(\"PRCA score\")\n",
    "fig.suptitle(\"CA score spread by transit group\")\n",
    "fig.tight_layout()\n",
    "fig.savefig(OUT / \"fig_ca_boxplots_by_transit.png\", dpi=150)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "15f414fb",
   "metadata": {},
   "source": [
    "## 7. Band prevalence (low / moderate / high)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "de8bd61c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:33.016934Z",
     "iopub.status.busy": "2026-07-25T15:42:33.016833Z",
     "iopub.status.idle": "2026-07-25T15:42:33.022357Z",
     "shell.execute_reply": "2026-07-25T15:42:33.021317Z"
    }
   },
   "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>band_col</th>\n",
       "      <th>group</th>\n",
       "      <th>n</th>\n",
       "      <th>low_n</th>\n",
       "      <th>low_pct</th>\n",
       "      <th>moderate_n</th>\n",
       "      <th>moderate_pct</th>\n",
       "      <th>high_n</th>\n",
       "      <th>high_pct</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>gt_group_band</td>\n",
       "      <td>overall</td>\n",
       "      <td>241</td>\n",
       "      <td>130</td>\n",
       "      <td>0.5394</td>\n",
       "      <td>61</td>\n",
       "      <td>0.2531</td>\n",
       "      <td>50</td>\n",
       "      <td>0.2075</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>gt_group_band</td>\n",
       "      <td>regular</td>\n",
       "      <td>101</td>\n",
       "      <td>65</td>\n",
       "      <td>0.6436</td>\n",
       "      <td>23</td>\n",
       "      <td>0.2277</td>\n",
       "      <td>13</td>\n",
       "      <td>0.1287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>gt_group_band</td>\n",
       "      <td>not_regular</td>\n",
       "      <td>140</td>\n",
       "      <td>65</td>\n",
       "      <td>0.4643</td>\n",
       "      <td>38</td>\n",
       "      <td>0.2714</td>\n",
       "      <td>37</td>\n",
       "      <td>0.2643</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>gt_interpersonal_band</td>\n",
       "      <td>overall</td>\n",
       "      <td>241</td>\n",
       "      <td>128</td>\n",
       "      <td>0.5311</td>\n",
       "      <td>69</td>\n",
       "      <td>0.2863</td>\n",
       "      <td>44</td>\n",
       "      <td>0.1826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>gt_interpersonal_band</td>\n",
       "      <td>regular</td>\n",
       "      <td>101</td>\n",
       "      <td>63</td>\n",
       "      <td>0.6238</td>\n",
       "      <td>25</td>\n",
       "      <td>0.2475</td>\n",
       "      <td>13</td>\n",
       "      <td>0.1287</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>gt_interpersonal_band</td>\n",
       "      <td>not_regular</td>\n",
       "      <td>140</td>\n",
       "      <td>65</td>\n",
       "      <td>0.4643</td>\n",
       "      <td>44</td>\n",
       "      <td>0.3143</td>\n",
       "      <td>31</td>\n",
       "      <td>0.2214</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                band_col        group    n  low_n  low_pct  moderate_n  \\\n",
       "0          gt_group_band      overall  241    130   0.5394          61   \n",
       "1          gt_group_band      regular  101     65   0.6436          23   \n",
       "2          gt_group_band  not_regular  140     65   0.4643          38   \n",
       "3  gt_interpersonal_band      overall  241    128   0.5311          69   \n",
       "4  gt_interpersonal_band      regular  101     63   0.6238          25   \n",
       "5  gt_interpersonal_band  not_regular  140     65   0.4643          44   \n",
       "\n",
       "   moderate_pct  high_n  high_pct  \n",
       "0        0.2531      50    0.2075  \n",
       "1        0.2277      13    0.1287  \n",
       "2        0.2714      37    0.2643  \n",
       "3        0.2863      44    0.1826  \n",
       "4        0.2475      13    0.1287  \n",
       "5        0.3143      31    0.2214  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "analysis[\"bands\"]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a4e36832",
   "metadata": {},
   "source": [
    "## 8. Sensitivity — alternate definitions of “regular”\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "3029c8af",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:33.024081Z",
     "iopub.status.busy": "2026-07-25T15:42:33.023959Z",
     "iopub.status.idle": "2026-07-25T15:42:33.030988Z",
     "shell.execute_reply": "2026-07-25T15:42:33.030429Z"
    }
   },
   "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>cutoff</th>\n",
       "      <th>score</th>\n",
       "      <th>n_regular</th>\n",
       "      <th>n_not_regular</th>\n",
       "      <th>mean_regular</th>\n",
       "      <th>mean_not_regular</th>\n",
       "      <th>mean_overall</th>\n",
       "      <th>diff_regular_minus_not_regular</th>\n",
       "      <th>diff_regular_minus_overall</th>\n",
       "      <th>welch_p</th>\n",
       "      <th>cohens_d</th>\n",
       "      <th>significant_at_05</th>\n",
       "      <th>regular_labels</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>weekly_plus</td>\n",
       "      <td>gt_group_ca</td>\n",
       "      <td>101</td>\n",
       "      <td>140</td>\n",
       "      <td>13.0396</td>\n",
       "      <td>15.7643</td>\n",
       "      <td>14.6224</td>\n",
       "      <td>-2.7247</td>\n",
       "      <td>-1.5828</td>\n",
       "      <td>0.0003</td>\n",
       "      <td>-0.4624</td>\n",
       "      <td>True</td>\n",
       "      <td>4-8 days a month; 8 or more days a month</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>weekly_plus</td>\n",
       "      <td>gt_interpersonal_ca</td>\n",
       "      <td>101</td>\n",
       "      <td>140</td>\n",
       "      <td>13.3069</td>\n",
       "      <td>15.0357</td>\n",
       "      <td>14.3112</td>\n",
       "      <td>-1.7288</td>\n",
       "      <td>-1.0043</td>\n",
       "      <td>0.0201</td>\n",
       "      <td>-0.2998</td>\n",
       "      <td>True</td>\n",
       "      <td>4-8 days a month; 8 or more days a month</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>twice_monthly_plus</td>\n",
       "      <td>gt_group_ca</td>\n",
       "      <td>154</td>\n",
       "      <td>87</td>\n",
       "      <td>13.5455</td>\n",
       "      <td>16.5287</td>\n",
       "      <td>14.6224</td>\n",
       "      <td>-2.9833</td>\n",
       "      <td>-1.0770</td>\n",
       "      <td>0.0004</td>\n",
       "      <td>-0.5081</td>\n",
       "      <td>True</td>\n",
       "      <td>2-4 days a month; 4-8 days a month; 8 or more ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>twice_monthly_plus</td>\n",
       "      <td>gt_interpersonal_ca</td>\n",
       "      <td>154</td>\n",
       "      <td>87</td>\n",
       "      <td>13.4481</td>\n",
       "      <td>15.8391</td>\n",
       "      <td>14.3112</td>\n",
       "      <td>-2.3910</td>\n",
       "      <td>-0.8632</td>\n",
       "      <td>0.0032</td>\n",
       "      <td>-0.4184</td>\n",
       "      <td>True</td>\n",
       "      <td>2-4 days a month; 4-8 days a month; 8 or more ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>any_use</td>\n",
       "      <td>gt_group_ca</td>\n",
       "      <td>191</td>\n",
       "      <td>50</td>\n",
       "      <td>13.9005</td>\n",
       "      <td>17.3800</td>\n",
       "      <td>14.6224</td>\n",
       "      <td>-3.4795</td>\n",
       "      <td>-0.7219</td>\n",
       "      <td>0.0015</td>\n",
       "      <td>-0.5920</td>\n",
       "      <td>True</td>\n",
       "      <td>0-1 days a month; 2-4 days a month; 4-8 days a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>any_use</td>\n",
       "      <td>gt_interpersonal_ca</td>\n",
       "      <td>191</td>\n",
       "      <td>50</td>\n",
       "      <td>13.7801</td>\n",
       "      <td>16.3400</td>\n",
       "      <td>14.3112</td>\n",
       "      <td>-2.5599</td>\n",
       "      <td>-0.5311</td>\n",
       "      <td>0.0181</td>\n",
       "      <td>-0.4463</td>\n",
       "      <td>True</td>\n",
       "      <td>0-1 days a month; 2-4 days a month; 4-8 days a...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>heavy_only</td>\n",
       "      <td>gt_group_ca</td>\n",
       "      <td>55</td>\n",
       "      <td>186</td>\n",
       "      <td>12.8364</td>\n",
       "      <td>15.1505</td>\n",
       "      <td>14.6224</td>\n",
       "      <td>-2.3142</td>\n",
       "      <td>-1.7860</td>\n",
       "      <td>0.0070</td>\n",
       "      <td>-0.3879</td>\n",
       "      <td>True</td>\n",
       "      <td>8 or more days a month</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>heavy_only</td>\n",
       "      <td>gt_interpersonal_ca</td>\n",
       "      <td>55</td>\n",
       "      <td>186</td>\n",
       "      <td>13.1636</td>\n",
       "      <td>14.6505</td>\n",
       "      <td>14.3112</td>\n",
       "      <td>-1.4869</td>\n",
       "      <td>-1.1476</td>\n",
       "      <td>0.0894</td>\n",
       "      <td>-0.2565</td>\n",
       "      <td>False</td>\n",
       "      <td>8 or more days a month</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               cutoff                score  n_regular  n_not_regular  \\\n",
       "0         weekly_plus          gt_group_ca        101            140   \n",
       "1         weekly_plus  gt_interpersonal_ca        101            140   \n",
       "2  twice_monthly_plus          gt_group_ca        154             87   \n",
       "3  twice_monthly_plus  gt_interpersonal_ca        154             87   \n",
       "4             any_use          gt_group_ca        191             50   \n",
       "5             any_use  gt_interpersonal_ca        191             50   \n",
       "6          heavy_only          gt_group_ca         55            186   \n",
       "7          heavy_only  gt_interpersonal_ca         55            186   \n",
       "\n",
       "   mean_regular  mean_not_regular  mean_overall  \\\n",
       "0       13.0396           15.7643       14.6224   \n",
       "1       13.3069           15.0357       14.3112   \n",
       "2       13.5455           16.5287       14.6224   \n",
       "3       13.4481           15.8391       14.3112   \n",
       "4       13.9005           17.3800       14.6224   \n",
       "5       13.7801           16.3400       14.3112   \n",
       "6       12.8364           15.1505       14.6224   \n",
       "7       13.1636           14.6505       14.3112   \n",
       "\n",
       "   diff_regular_minus_not_regular  diff_regular_minus_overall  welch_p  \\\n",
       "0                         -2.7247                     -1.5828   0.0003   \n",
       "1                         -1.7288                     -1.0043   0.0201   \n",
       "2                         -2.9833                     -1.0770   0.0004   \n",
       "3                         -2.3910                     -0.8632   0.0032   \n",
       "4                         -3.4795                     -0.7219   0.0015   \n",
       "5                         -2.5599                     -0.5311   0.0181   \n",
       "6                         -2.3142                     -1.7860   0.0070   \n",
       "7                         -1.4869                     -1.1476   0.0894   \n",
       "\n",
       "   cohens_d  significant_at_05  \\\n",
       "0   -0.4624               True   \n",
       "1   -0.2998               True   \n",
       "2   -0.5081               True   \n",
       "3   -0.4184               True   \n",
       "4   -0.5920               True   \n",
       "5   -0.4463               True   \n",
       "6   -0.3879               True   \n",
       "7   -0.2565              False   \n",
       "\n",
       "                                      regular_labels  \n",
       "0           4-8 days a month; 8 or more days a month  \n",
       "1           4-8 days a month; 8 or more days a month  \n",
       "2  2-4 days a month; 4-8 days a month; 8 or more ...  \n",
       "3  2-4 days a month; 4-8 days a month; 8 or more ...  \n",
       "4  0-1 days a month; 2-4 days a month; 4-8 days a...  \n",
       "5  0-1 days a month; 2-4 days a month; 4-8 days a...  \n",
       "6                             8 or more days a month  \n",
       "7                             8 or more days a month  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sens = analysis[\"sensitivity\"]\n",
    "sens[[\n",
    "    \"cutoff\", \"score\", \"n_regular\", \"n_not_regular\",\n",
    "    \"mean_regular\", \"mean_not_regular\", \"mean_overall\",\n",
    "    \"diff_regular_minus_not_regular\", \"diff_regular_minus_overall\",\n",
    "    \"welch_p\", \"cohens_d\", \"significant_at_05\", \"regular_labels\",\n",
    "]]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "963ae4db",
   "metadata": {},
   "source": [
    "## 9. Write distributable artifacts\n",
    "\n",
    "Tables + JSON results card land in `outputs/transit_ca/` (gitignored). Use these for the manuscript / slides.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "7752c8ba",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:33.032076Z",
     "iopub.status.busy": "2026-07-25T15:42:33.031954Z",
     "iopub.status.idle": "2026-07-25T15:42:33.045249Z",
     "shell.execute_reply": "2026-07-25T15:42:33.044842Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'descriptives': '/workspace/outputs/transit_ca/transit_ca_descriptives.csv',\n",
       " 'comparisons': '/workspace/outputs/transit_ca/transit_ca_comparisons.csv',\n",
       " 'bands': '/workspace/outputs/transit_ca/transit_ca_band_prevalence.csv',\n",
       " 'by_q26': '/workspace/outputs/transit_ca/transit_ca_by_q26.csv',\n",
       " 'sensitivity': '/workspace/outputs/transit_ca/transit_ca_sensitivity.csv',\n",
       " 'distribution': '/workspace/outputs/transit_ca/transit_ca_score_distribution.csv',\n",
       " 'labeled': '/workspace/outputs/transit_ca/transit_ca_labeled_sample.csv',\n",
       " 'summary': '/workspace/outputs/transit_ca/transit_ca_summary.json',\n",
       " 'results_card': '/workspace/outputs/transit_ca/transit_ca_results_card.json'}"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "paths = save_transit_ca_artifacts(analysis, OUT)\n",
    "# Also persist the cleaned analytic sample used here.\n",
    "processed = ROOT / \"data\" / \"processed\"\n",
    "processed.mkdir(parents=True, exist_ok=True)\n",
    "participants.to_csv(processed / \"participants_scored.csv\", index=False)\n",
    "(processed / \"cleaning_report.json\").write_text(json.dumps(cleaning_report, indent=2))\n",
    "{k: str(v) for k, v in paths.items()}\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fcbdfa3a",
   "metadata": {},
   "source": [
    "## 10. Takeaway (auto-filled from results card)\n",
    "\n",
    "The cell below prints the takeaway from `summary` / `transit_ca_summary.json` so this section never carries blank placeholders.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "46f22047",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-07-25T15:42:33.047211Z",
     "iopub.status.busy": "2026-07-25T15:42:33.047114Z",
     "iopub.status.idle": "2026-07-25T15:42:33.051253Z",
     "shell.execute_reply": "2026-07-25T15:42:33.050816Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Regular definition: Q26 in {4-8 days a month, 8 or more days a month} (weekly-or-more public transit)\n",
      "N regular / N comparison: 101 / 140\n",
      "Group CA: mean difference vs overall = -1.58 pts; Welch p = 0.0002925; d = -0.462 (small).\n",
      "Interpersonal CA: mean difference vs overall = -1.00 pts; Welch p = 0.02012; d = -0.300 (small).\n",
      "Conclusion: regular riders DO differ significantly from the larger cohort under α = .05; largest |d| magnitude is small.\n",
      "Interpretation: At least one PRCA subscale differs significantly (Welch t, α=.05) between regular and non-regular riders.\n"
     ]
    }
   ],
   "source": [
    "def _cohen_size(d: float) -> str:\n",
    "    ad = abs(d)\n",
    "    if ad < 0.2:\n",
    "        return \"negligible\"\n",
    "    if ad < 0.5:\n",
    "        return \"small\"\n",
    "    if ad < 0.8:\n",
    "        return \"medium\"\n",
    "    return \"large\"\n",
    "\n",
    "sample = summary[\"sample\"]\n",
    "print(\"Regular definition:\", sample[\"regular_definition\"])\n",
    "print(f\"N regular / N comparison: {sample['n_regular']} / {sample['n_not_regular']}\")\n",
    "for row in summary[\"primary_tests\"]:\n",
    "    label = \"Group CA\" if row[\"score\"] == \"gt_group_ca\" else \"Interpersonal CA\"\n",
    "    print(\n",
    "        f\"{label}: mean difference vs overall = {row['diff_regular_minus_overall']:+.2f} pts; \"\n",
    "        f\"Welch p = {row['welch_p']:.4g}; d = {row['cohens_d']:.3f} \"\n",
    "        f\"({_cohen_size(row['cohens_d'])}).\"\n",
    "    )\n",
    "any_sig = summary[\"verdict\"][\"any_subscale_significant_at_05\"]\n",
    "ds = [abs(r[\"cohens_d\"]) for r in summary[\"primary_tests\"]]\n",
    "mag = _cohen_size(max(ds)) if ds else \"unknown\"\n",
    "print(\n",
    "    \"Conclusion: regular riders \"\n",
    "    + (\"DO\" if any_sig else \"do NOT\")\n",
    "    + \" differ significantly from the larger cohort under α = .05; \"\n",
    "    f\"largest |d| magnitude is {mag}.\"\n",
    ")\n",
    "print(\"Interpretation:\", summary[\"verdict\"][\"interpretation\"])\n"
   ]
  }
 ],
 "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}