{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "6cbed08f",
   "metadata": {},
   "source": [
    "# Assessing Model 2\n",
    "\n",
    "In 2010, the American Civil Liberties Union (ACLU) of Northern California presented a report on jury selection in Alameda County, California. The report concluded that certain racial and **ethnic groups** are **underrepresented** among jury panelists in Alameda County, and suggested some reforms of the process by which eligible jurors are assigned to panels. As you saw in the previous section, trial jurors are selected from a larger jury panel. By law, the jury panel is supposed to be representative of the population of eligible jurors. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6f2792bc-20b9-4bfd-a37d-ec3c7a6b9da5",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:01.586835Z",
     "iopub.status.busy": "2026-08-31T06:01:01.586581Z",
     "iopub.status.idle": "2026-08-31T06:01:01.968586Z",
     "shell.execute_reply": "2026-08-31T06:01:01.968356Z"
    },
    "tags": [
     "hide-input",
     "thebe-init",
     "runnable"
    ]
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "plt.style.use('fivethirtyeight')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "60fe21d6",
   "metadata": {},
   "source": [
    "```{contents}\n",
    ":local:\n",
    ":depth: 2\n",
    "```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ac086cd4",
   "metadata": {},
   "source": [
    "## Alameda County Jury Panels\n",
    "\n",
    "The focus of the study by the ACLU of Northern California was the racial and ethnic composition of jury panels in Alameda County. The ACLU compiled data on the composition of the jury panels in **11** felony trials in Alameda County in the years 2009 and 2010. In those panels, the total number of people who reported for jury service was 1453. The ACLU gathered **demographic data** on all of these prospective jurors and compared that data with the composition of all eligible jurors in the county.\n",
    "\n",
    "The data are tabulated below in a table called `jury`. In each category, the first numerical value is the proportion of all eligible juror candidates in that category. The second value is the proportion of people in that category among those who appeared for the process of selection into the jury.\n",
    "\n",
    "The labels for the different categories are taken from the ACLU report, which says, “The category 'other' includes people who identified as mixed race without identifying one primary racial identity, and individuals who did not identify a race or ethnicity.” The label Asian/PI means “Asian or Pacific Islander.” The label Black/AA means “Black or African-American.”"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "387e9a6a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:01.970025Z",
     "iopub.status.busy": "2026-08-31T06:01:01.969918Z",
     "iopub.status.idle": "2026-08-31T06:01:01.974462Z",
     "shell.execute_reply": "2026-08-31T06:01:01.974203Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "   Ethnicity  Eligible  Panels\n0   Asian/PI      0.15    0.26\n1   Black/AA      0.18    0.08\n2  Caucasian      0.54    0.54\n3   Hispanic      0.12    0.08\n4      Other      0.01    0.04",
      "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>Ethnicity</th>\n      <th>Eligible</th>\n      <th>Panels</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Asian/PI</td>\n      <td>0.15</td>\n      <td>0.26</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Black/AA</td>\n      <td>0.18</td>\n      <td>0.08</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Caucasian</td>\n      <td>0.54</td>\n      <td>0.54</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Hispanic</td>\n      <td>0.12</td>\n      <td>0.08</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Other</td>\n      <td>0.01</td>\n      <td>0.04</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 2
    }
   ],
   "source": [
    "data = {\n",
    "        'Ethnicity': ['Asian/PI', 'Black/AA', 'Caucasian', 'Hispanic', 'Other'],\n",
    "        'Eligible': [0.15, 0.18, 0.54, 0.12, 0.01],\n",
    "        'Panels': [0.26, 0.08, 0.54, 0.08, 0.04]\n",
    "}\n",
    "\n",
    "jury = pd.DataFrame(data)\n",
    "\n",
    "jury"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "bab5de93",
   "metadata": {},
   "source": [
    "Some ethnicities are overrepresented and some are underrepresented on the jury panels in the study. In particular, panelists in the Black/AA and Hispanic categories are underrepresented. A bar chart is helpful for visualizing the differences."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "e98beee9",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:01.975644Z",
     "iopub.status.busy": "2026-08-31T06:01:01.975571Z",
     "iopub.status.idle": "2026-08-31T06:01:02.020781Z",
     "shell.execute_reply": "2026-08-31T06:01:02.020574Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "Text(0.5, 0, 'Proportion')"
     },
     "execution_count": 3
    },
    {
     "output_type": "display_data",
     "metadata": {},
     "data": {
      "text/plain": "<Figure size 600x500 with 1 Axes>",
      "image/png": "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"
     }
    }
   ],
   "source": [
    "jury.plot(x='Ethnicity', y=['Eligible', 'Panels'], kind='barh', figsize=(6, 5))\n",
    "plt.xlabel('Proportion')\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c44b0dad",
   "metadata": {},
   "source": [
    "## Data vs. Random\n",
    "\n",
    "What if we select a random sample of **1,453** people from the population of eligible jurors? Will the distribution of their ethnicities look like the distribution of the panels above?\n",
    "\n",
    "We can answer these questions by using sample_proportions and augmenting the jury table with a column of the proportions in our sample.\n",
    "\n",
    "Technical note: Random samples of prospective jurors would be selected **without replacement**. However, when the size of a sample is small relative to the size of the population, sampling without replacement resembles sampling with replacement; the proportions in the population don’t change much between draws. The population of eligible jurors in Alameda County is over a million, and compared to that, a sample size of about 1500 is quite small. We will therefore sample with replacement.\n",
    "\n",
    "In the cell below, we sample at random 1453 times from the distribution of eligible jurors, and display the distribution of the random sample along with the distributions of the eligible jurors and the panel in the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "5ba79d55-4284-48f6-8551-6f5c0209df8d",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.021923Z",
     "iopub.status.busy": "2026-08-31T06:01:02.021859Z",
     "iopub.status.idle": "2026-08-31T06:01:02.023507Z",
     "shell.execute_reply": "2026-08-31T06:01:02.023324Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "np.random.seed(42)\n",
    "def sample_proportions(sample_size, probabilities):\n",
    "    \"\"\"Return the proportion of random draws for each outcome in a distribution.\n",
    "    This function is similar to np.random.multinomial, but returns proportions\n",
    "    instead of counts.\n",
    "    Args:\n",
    "        ``sample_size``: The size of the sample to draw from the distribution.\n",
    "        ``probabilities``: An array of probabilities that forms a distribution.\n",
    "    Returns:\n",
    "        An array with the same length as ``probability`` that sums to 1.\n",
    "    \"\"\"\n",
    "    return np.random.multinomial(sample_size, probabilities) / sample_size\n",
    "    # return np.random.multinomial(sample_size, probabilities) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a8ffac68",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.024552Z",
     "iopub.status.busy": "2026-08-31T06:01:02.024497Z",
     "iopub.status.idle": "2026-08-31T06:01:02.027574Z",
     "shell.execute_reply": "2026-08-31T06:01:02.027393Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "   Ethnicity  Eligible  Panels  Random Sample\n0   Asian/PI      0.15    0.26       0.143840\n1   Black/AA      0.18    0.08       0.182381\n2  Caucasian      0.54    0.54       0.555403\n3   Hispanic      0.12    0.08       0.108741\n4      Other      0.01    0.04       0.009635",
      "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>Ethnicity</th>\n      <th>Eligible</th>\n      <th>Panels</th>\n      <th>Random Sample</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Asian/PI</td>\n      <td>0.15</td>\n      <td>0.26</td>\n      <td>0.143840</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Black/AA</td>\n      <td>0.18</td>\n      <td>0.08</td>\n      <td>0.182381</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Caucasian</td>\n      <td>0.54</td>\n      <td>0.54</td>\n      <td>0.555403</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Hispanic</td>\n      <td>0.12</td>\n      <td>0.08</td>\n      <td>0.108741</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Other</td>\n      <td>0.01</td>\n      <td>0.04</td>\n      <td>0.009635</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 5
    }
   ],
   "source": [
    "eligible_population = jury['Eligible']\n",
    "sample_distribution = sample_proportions(1453, eligible_population)\n",
    "jury['Random Sample'] = sample_distribution\n",
    "jury"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1e1a4379",
   "metadata": {},
   "source": [
    "The distribution of the random sample is quite close to the distribution of the eligible population, unlike the distribution of the panels. As always, it helps to visualize."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "0df72ca4",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.028558Z",
     "iopub.status.busy": "2026-08-31T06:01:02.028499Z",
     "iopub.status.idle": "2026-08-31T06:01:02.065555Z",
     "shell.execute_reply": "2026-08-31T06:01:02.065376Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "<Axes: ylabel='Ethnicity'>"
     },
     "execution_count": 6
    },
    {
     "output_type": "display_data",
     "metadata": {},
     "data": {
      "text/plain": "<Figure size 600x500 with 1 Axes>",
      "image/png": "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"
     }
    }
   ],
   "source": [
    "jury.plot('Ethnicity', kind='barh', figsize=(6, 5))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d688dcc0",
   "metadata": {},
   "source": [
    "The bar chart shows that the distribution of the random **sample** resembles the **eligible** population, but the **distribution** of the panels does not.\n",
    "\n",
    "To assess whether this observation is particular to one random sample or more general, we can simulate multiple panels under the model of random selection and see what the simulations predict. But we won’t be able to look at thousands of bar charts like the one above. We need a statistic that will help us assess whether or not the model or random selection is supported by the data."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "5e8c18ce",
   "metadata": {},
   "source": [
    "## Statistic: Distance between Distributions\n",
    "\n",
    "We know how to measure how different two numbers are: if the numbers are `x` and `y`, the distance between them is `| x - y |`. Now we have to quantify the distance between two distributions. For example, we have to measure the distance between the blue and gold distributions below."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "54a8c8cc",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.066630Z",
     "iopub.status.busy": "2026-08-31T06:01:02.066568Z",
     "iopub.status.idle": "2026-08-31T06:01:02.098210Z",
     "shell.execute_reply": "2026-08-31T06:01:02.098036Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "<Axes: ylabel='Ethnicity'>"
     },
     "execution_count": 7
    },
    {
     "output_type": "display_data",
     "metadata": {},
     "data": {
      "text/plain": "<Figure size 600x500 with 1 Axes>",
      "image/png": "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"
     }
    }
   ],
   "source": [
    "jury.drop('Random Sample', axis=1).plot('Ethnicity', kind='barh', figsize=(6, 5))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e93f30c4",
   "metadata": {},
   "source": [
    "For this, we will compute a quantity called the total variation distance between two distributions. The calculation is as an extension of how we find the distance between two numbers.\n",
    "\n",
    "To compute the total variation distance, we first find the difference between the two proportions in each category."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "09d50d3f",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.099229Z",
     "iopub.status.busy": "2026-08-31T06:01:02.099166Z",
     "iopub.status.idle": "2026-08-31T06:01:02.102247Z",
     "shell.execute_reply": "2026-08-31T06:01:02.102073Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "   Ethnicity  Eligible  Panels  Difference\n0   Asian/PI      0.15    0.26        0.11\n1   Black/AA      0.18    0.08       -0.10\n2  Caucasian      0.54    0.54        0.00\n3   Hispanic      0.12    0.08       -0.04\n4      Other      0.01    0.04        0.03",
      "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>Ethnicity</th>\n      <th>Eligible</th>\n      <th>Panels</th>\n      <th>Difference</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Asian/PI</td>\n      <td>0.15</td>\n      <td>0.26</td>\n      <td>0.11</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Black/AA</td>\n      <td>0.18</td>\n      <td>0.08</td>\n      <td>-0.10</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Caucasian</td>\n      <td>0.54</td>\n      <td>0.54</td>\n      <td>0.00</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Hispanic</td>\n      <td>0.12</td>\n      <td>0.08</td>\n      <td>-0.04</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Other</td>\n      <td>0.01</td>\n      <td>0.04</td>\n      <td>0.03</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 8
    }
   ],
   "source": [
    "jury = jury.drop('Random Sample', axis=1)\n",
    "\n",
    "jury['Difference'] = jury['Panels'] - jury['Eligible']\n",
    "\n",
    "jury"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aee6f094",
   "metadata": {},
   "source": [
    "Take a look at the column Difference and notice that the sum of its entries is 0: the positive entries add up to 0.14, exactly canceling the total of the negative entries which is -0.14.\n",
    "\n",
    "This is numerical evidence of the fact that in the bar chart, the gold bars exceed the blue bars by exactly as much as the blue bars exceed the gold. The proportions in each of the two columns Panels and Eligible add up to 1, and so the give-and-take between their entries must add up to 0.\n",
    "\n",
    "To avoid the cancellation, we drop the negative signs and then add all the entries. But this gives us two times the total of the positive entries (equivalently, two times the total of the negative entries, with the sign removed). We don’t need that doubling, so we divide the sum by 2."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "dbef3565",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.103190Z",
     "iopub.status.busy": "2026-08-31T06:01:02.103135Z",
     "iopub.status.idle": "2026-08-31T06:01:02.105879Z",
     "shell.execute_reply": "2026-08-31T06:01:02.105713Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "   Ethnicity  Eligible  Panels  Difference  Absolute Difference\n0   Asian/PI      0.15    0.26        0.11                 0.11\n1   Black/AA      0.18    0.08       -0.10                 0.10\n2  Caucasian      0.54    0.54        0.00                 0.00\n3   Hispanic      0.12    0.08       -0.04                 0.04\n4      Other      0.01    0.04        0.03                 0.03",
      "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>Ethnicity</th>\n      <th>Eligible</th>\n      <th>Panels</th>\n      <th>Difference</th>\n      <th>Absolute Difference</th>\n    </tr>\n  </thead>\n  <tbody>\n    <tr>\n      <th>0</th>\n      <td>Asian/PI</td>\n      <td>0.15</td>\n      <td>0.26</td>\n      <td>0.11</td>\n      <td>0.11</td>\n    </tr>\n    <tr>\n      <th>1</th>\n      <td>Black/AA</td>\n      <td>0.18</td>\n      <td>0.08</td>\n      <td>-0.10</td>\n      <td>0.10</td>\n    </tr>\n    <tr>\n      <th>2</th>\n      <td>Caucasian</td>\n      <td>0.54</td>\n      <td>0.54</td>\n      <td>0.00</td>\n      <td>0.00</td>\n    </tr>\n    <tr>\n      <th>3</th>\n      <td>Hispanic</td>\n      <td>0.12</td>\n      <td>0.08</td>\n      <td>-0.04</td>\n      <td>0.04</td>\n    </tr>\n    <tr>\n      <th>4</th>\n      <td>Other</td>\n      <td>0.01</td>\n      <td>0.04</td>\n      <td>0.03</td>\n      <td>0.03</td>\n    </tr>\n  </tbody>\n</table>\n</div>"
     },
     "execution_count": 9
    }
   ],
   "source": [
    "jury['Absolute Difference'] = np.abs(jury['Difference'])\n",
    "\n",
    "jury"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "a24b86dd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.106789Z",
     "iopub.status.busy": "2026-08-31T06:01:02.106737Z",
     "iopub.status.idle": "2026-08-31T06:01:02.108325Z",
     "shell.execute_reply": "2026-08-31T06:01:02.108143Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "np.float64(0.14)"
     },
     "execution_count": 10
    }
   ],
   "source": [
    "jury['Absolute Difference'].sum() / 2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "b4bcf1f5",
   "metadata": {},
   "source": [
    "This quantity 0.14 is the **total variation distance (TVD)** between the distribution of ethnicities in the eligible juror population and the distribution in the panels.\n",
    "\n",
    "In general, the total variation distance between two distributions measures how close the distributions are. The larger the TVD, the more different the two distributions appear.\n",
    "\n",
    "Technical Note: We could have obtained the same result by just adding the positive differences. But our method of including all the absolute differences eliminates the need to keep track of which differences are positive and which are not.\n",
    "\n",
    "We will use the total variation distance between distributions as the statistic to simulate under the assumption of random selection. Large values of the distance will be evidence against random selection.\n",
    "\n",
    "For example, a TVD of 0.14 means that 14% of the total probability mass would have to move between categories to make these distributions match. It does not mean a 14% probability that the selection process was biased.\n",
    "\n",
    "**Check your reasoning:** The signed differences add to zero because both distributions sum to one. Taking absolute values prevents cancellation. Dividing by two prevents counting both the amount removed from some categories and the same amount added to others.\n",
    "\n",
    "```{index} Total variation distance\n",
    "```\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3887e0ae",
   "metadata": {},
   "source": [
    "## Simulating the Statistic Under the Model\n",
    "\n",
    "To see how the TVD varies across random samples, we will simulate it repeatedly under the model of random selection from the eligible population. Let’s organize our calculation. Since we are going to be computing total variation distance repeatedly, we will first write a function that computes it for two given distributions.\n",
    "\n",
    "The function `total_variation_distance` takes two arrays containing the distributions to compare, and returns the TVD between them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "eab0c661",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.109265Z",
     "iopub.status.busy": "2026-08-31T06:01:02.109209Z",
     "iopub.status.idle": "2026-08-31T06:01:02.110490Z",
     "shell.execute_reply": "2026-08-31T06:01:02.110313Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "def total_variation_distance(distribution_1, distribution_2):\n",
    "    return sum(np.abs(distribution_1 - distribution_2)) / 2"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e0b94efa",
   "metadata": {},
   "source": [
    "This function will help us calculate our statistic in each repetition of the simulation. But first let’s check that it gives the right answer when we use it to compute the distance between the blue (eligible) and gold (panels) distributions above. These are the distribution in the ACLU study.\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "e8b616cd",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.111464Z",
     "iopub.status.busy": "2026-08-31T06:01:02.111400Z",
     "iopub.status.idle": "2026-08-31T06:01:02.112962Z",
     "shell.execute_reply": "2026-08-31T06:01:02.112814Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "0.14"
     },
     "execution_count": 12
    }
   ],
   "source": [
    "total_variation_distance( jury['Panels'], jury['Eligible'] )"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3e9774ce",
   "metadata": {},
   "source": [
    "This agrees with the value that we computed directly without using the function.\n",
    "\n",
    "In the cell below, we use the function to compute the **TVD** between the **distributions** of the eligible jurors and one random sample. Recall that eligible_population is the array containing the distribution of the eligible jurors, and that our sample size is 1453.\n",
    "\n",
    "In the first line, we use sample_proportions to generate a random sample from the eligible population. In the next line, we use total_variation_distance to compute the TVD between the distributions in the random sample and the eligible population."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "481a5e90",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.113840Z",
     "iopub.status.busy": "2026-08-31T06:01:02.113797Z",
     "iopub.status.idle": "2026-08-31T06:01:02.115402Z",
     "shell.execute_reply": "2026-08-31T06:01:02.115263Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "0.024845147969717856"
     },
     "execution_count": 13
    }
   ],
   "source": [
    "sample_distribution = sample_proportions(1453, eligible_population)\n",
    "total_variation_distance(sample_distribution, eligible_population)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "43f25431",
   "metadata": {},
   "source": [
    "Run the cell a few times and notice that the distances are quite a bit smaller than 0.14, the distance between the distribution of the panels and the eligible jurors.\n",
    "\n",
    "We are now ready to run a simulation to assess the model of random selection."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "265a7047",
   "metadata": {},
   "source": [
    "### Simulating One Value of the Statistic\n",
    "\n",
    "In the same way that we start every simulation, let’s define a function one_simulated_tvd that returns one simulated value of the total variation distance under the hypothesis of random selection.\n",
    "\n",
    "The code in the body of the definition is based on the cell above."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "e6a2b0ea",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.116274Z",
     "iopub.status.busy": "2026-08-31T06:01:02.116233Z",
     "iopub.status.idle": "2026-08-31T06:01:02.117475Z",
     "shell.execute_reply": "2026-08-31T06:01:02.117325Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Simulate one simulated value of \n",
    "# the total variation distance between\n",
    "# the distribution of a sample selected at random\n",
    "# and the distribution of the eligible population\n",
    "\n",
    "def one_simulated_tvd():\n",
    "    sample_distribution = sample_proportions(1453, eligible_population)\n",
    "    return total_variation_distance(sample_distribution, eligible_population)   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "tags": [
     "thebe-interactive"
    ]
   },
   "outputs": [],
   "source": [
    "### Exercise: Simulate One TVD\n",
    "# A store's customers are 40% North, 35% Central, and 25% South.\n",
    "# Simulate a random sample of 100 customers. Convert counts to proportions,\n",
    "# then print the TVD from the population shares, rounded to 3 decimals.\n",
    "import numpy as np\n",
    "rng = np.random.default_rng(17)\n",
    "population = np.array([0.40, 0.35, 0.25])\n",
    "### Your code starts here.\n",
    "\n",
    "### Your code ends here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "tags": [
     "hide-input"
    ]
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": "TVD: 0.13\n"
    }
   ],
   "source": [
    "### Solution\n",
    "import numpy as np\n",
    "rng = np.random.default_rng(17)\n",
    "population = np.array([0.40, 0.35, 0.25])\n",
    "sample = rng.multinomial(100, population) / 100\n",
    "print('TVD:', round(np.abs(sample - population).sum() / 2, 3))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b030bda",
   "metadata": {},
   "source": [
    "### Simulating Multiple Values\n",
    "\n",
    "Now we can apply the familiar process of using a for loop to create an array consisting of 5000 such distances."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "776cc46c",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.118320Z",
     "iopub.status.busy": "2026-08-31T06:01:02.118280Z",
     "iopub.status.idle": "2026-08-31T06:01:02.250965Z",
     "shell.execute_reply": "2026-08-31T06:01:02.250685Z"
    },
    "tags": []
   },
   "outputs": [],
   "source": [
    "tvds = np.array([])\n",
    "repetitions = 5000\n",
    "for i in np.arange(repetitions):\n",
    "    tvds = np.append(tvds, one_simulated_tvd())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "tags": [
     "thebe-interactive"
    ]
   },
   "outputs": [],
   "source": [
    "### Exercise: Repeat the Model\n",
    "# Complete 100 repetitions of the store model. Each repetition must draw\n",
    "# a new sample of 100 customers and record its TVD from population.\n",
    "# Print the number of statistics collected.\n",
    "import numpy as np\n",
    "rng = np.random.default_rng(17)\n",
    "population = np.array([0.40, 0.35, 0.25])\n",
    "tvds_practice = []\n",
    "### Your code starts here.\n",
    "\n",
    "### Your code ends here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "tags": [
     "hide-input"
    ]
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": "Statistics collected: 100\n"
    }
   ],
   "source": [
    "### Solution\n",
    "import numpy as np\n",
    "rng = np.random.default_rng(17)\n",
    "population = np.array([0.40, 0.35, 0.25])\n",
    "tvds_practice = []\n",
    "for _ in range(100):\n",
    "    sample = rng.multinomial(100, population) / 100\n",
    "    tvds_practice.append(np.abs(sample - population).sum() / 2)\n",
    "print('Statistics collected:', len(tvds_practice))\n",
    "# Repeating one saved sample would not create a null distribution."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "48df9b13",
   "metadata": {},
   "source": [
    "## Assessing the Model of Random Selection\n",
    "\n",
    "Here is the empirical histogram of the simulated distances. It shows that if you draw 1453 panelists at random from the pool of eligible candidates, then the distance between the distributions of the panelists and the eligible population is rarely more than about 0.05.\n",
    "\n",
    "The panels in the study, however, were not quite so similar to the eligible population. The total variation distance between the panels and the population was 0.14, shown as the red dashed line. It is far beyond the tail of the histogram and does not look at all like a typical distance between the distributions of a random sample and the eligible population."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "60349229",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.252104Z",
     "iopub.status.busy": "2026-08-31T06:01:02.252047Z",
     "iopub.status.idle": "2026-08-31T06:01:02.299971Z",
     "shell.execute_reply": "2026-08-31T06:01:02.299763Z"
    },
    "tags": []
   },
   "outputs": [
    {
     "output_type": "execute_result",
     "metadata": {},
     "data": {
      "text/plain": "<matplotlib.legend.Legend at 0x112635810>"
     },
     "execution_count": 20
    },
    {
     "output_type": "display_data",
     "metadata": {},
     "data": {
      "text/plain": "<Figure size 700x400 with 1 Axes>",
      "image/png": "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"
     }
    }
   ],
   "source": [
    "fig, ax = plt.subplots(figsize=(7, 4))\n",
    "ax.hist(tvds, bins=np.arange(0, 0.205, 0.005), density=True, edgecolor='white')\n",
    "ax.axvspan(0.14, 0.20, color='red', alpha=0.15, label='At least the observed TVD')\n",
    "ax.axvline(0.14, color='red', linestyle='--', label='Observed TVD: 0.14')\n",
    "ax.set_title('Prediction Assuming Random Selection')\n",
    "ax.set_xlabel('Total Variation Distance')\n",
    "ax.set_ylabel('Probability Density')\n",
    "ax.set_xlim(0, 0.20)\n",
    "ax.legend()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a52974e8",
   "metadata": {},
   "source": [
    "The simulation shows that the composition of the panels in the ACLU study is not consistent with the model of random selection. Our analysis supports the ACLU’s conclusion that the panels were not representative of the distribution provided for the eligible jurors.\n",
    "\n",
    "The shaded region marks the direction of evidence against the model: TVDs at least as large as 0.14. With equal-width histogram bins, its area corresponds to the proportion of simulated results in that tail. Density is height per unit of TVD; the height of an individual bar is not itself a probability.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "aa753c00",
   "metadata": {},
   "source": [
    "## Conclusion of the Data Analysis\n",
    "\n",
    "### Data Quality\n",
    "\n",
    "We have developed a technique that helps us decide whether one distribution looks like a random sample from another. But we should examine our data sources to better understand exactly what we can conclude from our analysis. Good data science includes a thoughtful examination of how the data were gathered. For Case 2:\n",
    "\n",
    "**Eligible Jurors**. It is important to keep in mind that not everyone is eligible to serve on a jury. The historical case describes the eligibility criteria used at the time as follows: “You may be called to serve if you are 18 years old, a U.S. citizen and a resident of the county or district where summoned. You must be able to understand English, and be physically and mentally capable of serving. In addition, you must not have served as any kind of juror in the past 12 months, nor have been convicted of a felony.”\n",
    "\n",
    "Though the Census is a natural source of demographic information, the Census doesn’t collect data on all these categories. Thus the ACLU had to obtain the demographics of eligible jurors in some other way.\n",
    "\n",
    "What they used was a set of estimates that had been developed by a San Diego State University professor for an Alameda County trial in 2002. Those estimates were based on the 2000 Census and also took into account the criteria required for eligibility as a juror.\n",
    "\n",
    "So we have to keep in mind that the distribution that we used for the eligible population is itself an estimate and is therefore subject to error. Moreover, as the ACLU noted, using estimates based on the 2000 Census for populations in 2010 might not be accurate due to the changing demographics in California.\n",
    "\n",
    "Thus the distribution used for the population of eligible jurors might not be accurate.\n",
    "\n",
    "**Panels**. It is not clear from the report exactly how the 1453 panelists were classified into the different ethnic categories. The report says only that “attorneys … cooperated in collecting jury pool data”.\n",
    "\n",
    "Significant social, cultural, and political factors affect who gets classified or self-classifies into each ethnic category. It is also not clear if the definitions of those categories in the panels were the same as those used in the estimates of the distribution of eligible jurors. This raises questions about the correspondence between the two distributions being compared.\n",
    "\n",
    "Moreover, the data on panelists was obtained from those who reported for service. As we have discussed, not all panelists do so. The reasons for not reporting are associated with race and ethnicity, and disproportionately affect panelists from under-resourced communities.\n",
    "\n",
    "Responding to jury summons is required by law. But as the ACLU noted, at the time of the study Alameda County did not have an effective process for following up on prospective jurors who had been called but had failed to appear. The ACLU recommended the creation of “a failure to appear program that requires, at a minimum, the issuance of a second notice to prospective jurors.”\n",
    "\n",
    "Indeed, following up on non-respondents would reduce bias. For our analysis, we must acknowledge that we don’t have the data needed to fill in the demographic information for panelists who were selected but did not report for service.\n",
    "\n",
    "**Check your reasoning:** A simulation assumes the supplied population shares are correct. Increasing the number of repetitions reduces simulation noise, but it cannot repair outdated population estimates, inconsistent classifications, or missing nonrespondents. Before recommending a process change, investigate those data limitations.\n",
    "\n",
    "```{index} Data quality\n",
    "```\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "tags": [
     "thebe-interactive"
    ]
   },
   "outputs": [],
   "source": [
    "### Exercise: Check the Data Source\n",
    "# A store uses customer shares from ten years ago as its current population.\n",
    "# Would increasing a simulation from 100 to 10,000 repetitions fix this issue?\n",
    "# Print yes or no and explain your answer in a comment.\n",
    "### Your code starts here.\n",
    "\n",
    "### Your code ends here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "tags": [
     "hide-input"
    ]
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": "no\n"
    }
   ],
   "source": [
    "### Solution\n",
    "print('no')\n",
    "# More repetitions improve the simulation of the supplied model. They do not\n",
    "# establish that old customer shares describe the current population."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cell-f52f9ffc",
   "metadata": {},
   "source": [
    "\n",
    "## Section Practice\n",
    "\n",
    "Use this short exercise to check the main workflow from this section.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "cell-1f22d1e1",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.301083Z",
     "iopub.status.busy": "2026-08-31T06:01:02.301029Z",
     "iopub.status.idle": "2026-08-31T06:01:02.302353Z",
     "shell.execute_reply": "2026-08-31T06:01:02.302164Z"
    },
    "tags": [
     "thebe-interactive"
    ]
   },
   "outputs": [],
   "source": [
    "### Exercise: Practice Total Variation Distance\n",
    "# A company compares the department mix of this year's new hires with the mix of\n",
    "# all applicants.\n",
    "#   1. Compute the absolute difference between the two distributions for each department.\n",
    "#   2. Compute the total variation distance (TVD): half the sum of the absolute differences.\n",
    "#   3. Print the TVD rounded to 2 decimals.\n",
    "import pandas as pd\n",
    "\n",
    "mix = pd.DataFrame({\n",
    "    'department': ['Sales', 'Operations', 'Finance', 'IT'],\n",
    "    'applicants': [0.35, 0.30, 0.15, 0.20],\n",
    "    'hires': [0.25, 0.30, 0.20, 0.25],\n",
    "})\n",
    "### Your code starts here\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "### Your code ends here\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "cell-4562f86a",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2026-08-31T06:01:02.303254Z",
     "iopub.status.busy": "2026-08-31T06:01:02.303198Z",
     "iopub.status.idle": "2026-08-31T06:01:02.304682Z",
     "shell.execute_reply": "2026-08-31T06:01:02.304513Z"
    },
    "tags": [
     "hide-input"
    ]
   },
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": "   department  applicants  hires  abs_diff\n0       Sales        0.35   0.25      0.10\n1  Operations        0.30   0.30      0.00\n2     Finance        0.15   0.20      0.05\n3          IT        0.20   0.25      0.05\nTVD: 0.1\n"
    }
   ],
   "source": [
    "### Solution\n",
    "import pandas as pd\n",
    "\n",
    "mix = pd.DataFrame({\n",
    "    'department': ['Sales', 'Operations', 'Finance', 'IT'],\n",
    "    'applicants': [0.35, 0.30, 0.15, 0.20],\n",
    "    'hires': [0.25, 0.30, 0.20, 0.25],\n",
    "})\n",
    "\n",
    "mix['abs_diff'] = (mix['hires'] - mix['applicants']).abs()\n",
    "tvd = mix['abs_diff'].sum() / 2\n",
    "\n",
    "print(mix)\n",
    "print('TVD:', round(tvd, 2))\n",
    "# To judge whether 0.10 is large, compare it with TVDs simulated from random hiring.\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.13.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}