Skip to content

Instantly share code, notes, and snippets.

@guinetn

guinetn/ascombe-quartet-message.ipynb

Created August 14, 2022 12:33
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save guinetn/d4f108b822731a764e04bfb6bf818fcd to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save guinetn/d4f108b822731a764e04bfb6bf818fcd to your computer and use it in GitHub Desktop.
Download ZIP
Anscombe's quartet: Always Plot Your Data Before Jumping to Conclusions !
Display the source blob
Display the rendered blob
Raw
ascombe-quartet-message.ipynb
{
"cells": [
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# Anscombe's quartet\n",
"\n",
"### Abstract\n",
"\n",
"📌 **You Should Always Plot Your Data Before Jumping to Conclusions** \n",
"Demonstrate both the importance of graphing data when analyzing data, and the effect of outliers and other influential observations on statistical properties. \n",
"\n",
"### Keys\n",
"\n",
"- 📌 Same stats but different graphs! \n",
"- “Even if you’re very good at statistics, you might miss something”\n",
"\n",
"### Materials\n",
"\n",
"Anscombe's quartet (aka The Troll Dataset)\n",
"- Well-known statistical dataset trick with a powerful message.\n",
"it shows data visualization is not an optional or inferior angle to analytics. What we may have totally missed with the summary statistics was clear as day with the visualization.\n",
"- Dataset created by statistician Francis Anscome in 1973. \n",
"- 4 sets of eleven pairs of (x,y) coordinates data (labelled with cool Roman numerals I, II, III, IV)\n",
"- We can explored and compared it in two ways: summarize or visualize it.\n",
"\n"
]
},
{
"cell_type": "code",
"execution_count": 43,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr style=\"text-align: right;\">\n",
" <th></th>\n",
" <th>dataset</th>\n",
" <th>x</th>\n",
" <th>y</th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>0</th>\n",
" <td>I</td>\n",
" <td>10.0</td>\n",
" <td>8.04</td>\n",
" </tr>\n",
" <tr>\n",
" <th>1</th>\n",
" <td>I</td>\n",
" <td>8.0</td>\n",
" <td>6.95</td>\n",
" </tr>\n",
" <tr>\n",
" <th>2</th>\n",
" <td>I</td>\n",
" <td>13.0</td>\n",
" <td>7.58</td>\n",
" </tr>\n",
" <tr>\n",
" <th>3</th>\n",
" <td>I</td>\n",
" <td>9.0</td>\n",
" <td>8.81</td>\n",
" </tr>\n",
" <tr>\n",
" <th>4</th>\n",
" <td>I</td>\n",
" <td>11.0</td>\n",
" <td>8.33</td>\n",
" </tr>\n",
" <tr>\n",
" <th>5</th>\n",
" <td>I</td>\n",
" <td>14.0</td>\n",
" <td>9.96</td>\n",
" </tr>\n",
" <tr>\n",
" <th>6</th>\n",
" <td>I</td>\n",
" <td>6.0</td>\n",
" <td>7.24</td>\n",
" </tr>\n",
" <tr>\n",
" <th>7</th>\n",
" <td>I</td>\n",
" <td>4.0</td>\n",
" <td>4.26</td>\n",
" </tr>\n",
" <tr>\n",
" <th>8</th>\n",
" <td>I</td>\n",
" <td>12.0</td>\n",
" <td>10.84</td>\n",
" </tr>\n",
" <tr>\n",
" <th>9</th>\n",
" <td>I</td>\n",
" <td>7.0</td>\n",
" <td>4.82</td>\n",
" </tr>\n",
" <tr>\n",
" <th>10</th>\n",
" <td>I</td>\n",
" <td>5.0</td>\n",
" <td>5.68</td>\n",
" </tr>\n",
" <tr>\n",
" <th>11</th>\n",
" <td>II</td>\n",
" <td>10.0</td>\n",
" <td>9.14</td>\n",
" </tr>\n",
" <tr>\n",
" <th>12</th>\n",
" <td>II</td>\n",
" <td>8.0</td>\n",
" <td>8.14</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" dataset x y\n",
"0 I 10.0 8.04\n",
"1 I 8.0 6.95\n",
"2 I 13.0 7.58\n",
"3 I 9.0 8.81\n",
"4 I 11.0 8.33\n",
"5 I 14.0 9.96\n",
"6 I 6.0 7.24\n",
"7 I 4.0 4.26\n",
"8 I 12.0 10.84\n",
"9 I 7.0 4.82\n",
"10 I 5.0 5.68\n",
"11 II 10.0 9.14\n",
"12 II 8.0 8.14"
]
},
"execution_count": 43,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"import seaborn as sns\n",
"import pandas as pd\n",
"anscombe = sns.load_dataset('anscombe')\n",
"anscombe.head(13) # 13 to show the 'II' dataset starting..."
]
},
{
"cell_type": "code",
"execution_count": 32,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"(44, 1)"
]
},
"execution_count": 32,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Pretty simple dataset...\n",
"anscombe.shape"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# A. Dataset summary statistics\n",
"\n",
"See differences? Not a lot. Very similar groups."
]
},
{
"cell_type": "code",
"execution_count": 10,
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"<div>\n",
"<style scoped>\n",
" .dataframe tbody tr th:only-of-type {\n",
" vertical-align: middle;\n",
" }\n",
"\n",
" .dataframe tbody tr th {\n",
" vertical-align: top;\n",
" }\n",
"\n",
" .dataframe thead tr th {\n",
" text-align: left;\n",
" }\n",
"\n",
" .dataframe thead tr:last-of-type th {\n",
" text-align: right;\n",
" }\n",
"</style>\n",
"<table border=\"1\" class=\"dataframe\">\n",
" <thead>\n",
" <tr>\n",
" <th></th>\n",
" <th colspan=\"8\" halign=\"left\">x</th>\n",
" <th colspan=\"8\" halign=\"left\">y</th>\n",
" </tr>\n",
" <tr>\n",
" <th></th>\n",
" <th>count</th>\n",
" <th>mean</th>\n",
" <th>std</th>\n",
" <th>min</th>\n",
" <th>25%</th>\n",
" <th>50%</th>\n",
" <th>75%</th>\n",
" <th>max</th>\n",
" <th>count</th>\n",
" <th>mean</th>\n",
" <th>std</th>\n",
" <th>min</th>\n",
" <th>25%</th>\n",
" <th>50%</th>\n",
" <th>75%</th>\n",
" <th>max</th>\n",
" </tr>\n",
" <tr>\n",
" <th>dataset</th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" <th></th>\n",
" </tr>\n",
" </thead>\n",
" <tbody>\n",
" <tr>\n",
" <th>I</th>\n",
" <td>11.0</td>\n",
" <td>9.0</td>\n",
" <td>3.316625</td>\n",
" <td>4.0</td>\n",
" <td>6.5</td>\n",
" <td>9.0</td>\n",
" <td>11.5</td>\n",
" <td>14.0</td>\n",
" <td>11.0</td>\n",
" <td>7.500909</td>\n",
" <td>2.031568</td>\n",
" <td>4.26</td>\n",
" <td>6.315</td>\n",
" <td>7.58</td>\n",
" <td>8.57</td>\n",
" <td>10.84</td>\n",
" </tr>\n",
" <tr>\n",
" <th>II</th>\n",
" <td>11.0</td>\n",
" <td>9.0</td>\n",
" <td>3.316625</td>\n",
" <td>4.0</td>\n",
" <td>6.5</td>\n",
" <td>9.0</td>\n",
" <td>11.5</td>\n",
" <td>14.0</td>\n",
" <td>11.0</td>\n",
" <td>7.500909</td>\n",
" <td>2.031657</td>\n",
" <td>3.10</td>\n",
" <td>6.695</td>\n",
" <td>8.14</td>\n",
" <td>8.95</td>\n",
" <td>9.26</td>\n",
" </tr>\n",
" <tr>\n",
" <th>III</th>\n",
" <td>11.0</td>\n",
" <td>9.0</td>\n",
" <td>3.316625</td>\n",
" <td>4.0</td>\n",
" <td>6.5</td>\n",
" <td>9.0</td>\n",
" <td>11.5</td>\n",
" <td>14.0</td>\n",
" <td>11.0</td>\n",
" <td>7.500000</td>\n",
" <td>2.030424</td>\n",
" <td>5.39</td>\n",
" <td>6.250</td>\n",
" <td>7.11</td>\n",
" <td>7.98</td>\n",
" <td>12.74</td>\n",
" </tr>\n",
" <tr>\n",
" <th>IV</th>\n",
" <td>11.0</td>\n",
" <td>9.0</td>\n",
" <td>3.316625</td>\n",
" <td>8.0</td>\n",
" <td>8.0</td>\n",
" <td>8.0</td>\n",
" <td>8.0</td>\n",
" <td>19.0</td>\n",
" <td>11.0</td>\n",
" <td>7.500909</td>\n",
" <td>2.030579</td>\n",
" <td>5.25</td>\n",
" <td>6.170</td>\n",
" <td>7.04</td>\n",
" <td>8.19</td>\n",
" <td>12.50</td>\n",
" </tr>\n",
" </tbody>\n",
"</table>\n",
"</div>"
],
"text/plain": [
" x y \\\n",
" count mean std min 25% 50% 75% max count mean \n",
"dataset \n",
"I 11.0 9.0 3.316625 4.0 6.5 9.0 11.5 14.0 11.0 7.500909 \n",
"II 11.0 9.0 3.316625 4.0 6.5 9.0 11.5 14.0 11.0 7.500909 \n",
"III 11.0 9.0 3.316625 4.0 6.5 9.0 11.5 14.0 11.0 7.500000 \n",
"IV 11.0 9.0 3.316625 8.0 8.0 8.0 8.0 19.0 11.0 7.500909 \n",
"\n",
" \n",
" std min 25% 50% 75% max \n",
"dataset \n",
"I 2.031568 4.26 6.315 7.58 8.57 10.84 \n",
"II 2.031657 3.10 6.695 8.14 8.95 9.26 \n",
"III 2.030424 5.39 6.250 7.11 7.98 12.74 \n",
"IV 2.030579 5.25 6.170 7.04 8.19 12.50 "
]
},
"execution_count": 10,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"anscombe.groupby('dataset').describe()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"# B. Dataset visualization\n",
"\n",
"Because the human brain is rubbish at gleaning information from tabular data — even for this pathetically small dataset — let's plot out each of the four datasets and compare them.\n",
"\n",
"All four sets are identical when examined using simple summary statistics, but vary considerably when graphed:\n"
]
},
{
"cell_type": "code",
"execution_count": 58,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<seaborn.axisgrid.FacetGrid at 0x1baed036fa0>"
]
},
"execution_count": 58,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 864x216 with 4 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"\n",
"g = sns.FacetGrid(anscombe, col=\"dataset\")\n",
"g.map(sns.scatterplot, \"x\", \"y\")"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Another point of view:"
]
},
{
"cell_type": "code",
"execution_count": 44,
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"<AxesSubplot:xlabel='x', ylabel='y'>"
]
},
"execution_count": 44,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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",
"text/plain": [
"<Figure size 576x396 with 1 Axes>"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"sns.scatterplot(data=anscombe, x='x', y='y', hue='dataset')"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"The pairs looked nearly identical when we summarized them, but the visualization shows great differences. Thanks outliers!\n",
"\n",
"The group of four data sets that have the same “summary statistics” (mean, standard deviation, Pearson’s correlation). Yet they each produce wildly different graphs. It’s a famed demonstration of just how vital it can be to visualize data rather than relying on statistics alone\n",
"\n",
"# Moral of the story\n",
"\n",
"Anscombe’s quartet show us that data visualization is not an optional or inferior angle to analytics. It is a part of the analytic process. \n",
"Secondary moral is to be very skeptical of small sample sizes.\n",
"\n",
"# Generalization\n",
"\n",
"Since its publication, several methods to generate similar data sets with identical statistics and dissimilar graphics have been developed. One of these, the Datasaurus Dozen, consists of points tracing out the outline of a dinosaur, plus twelve other data sets that have the same summary statistics"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"### More\n",
"\n",
"- https://en.wikipedia.org/wiki/Anscombe%27s_quartet\n",
"- https://blog.revolutionanalytics.com/2017/05/the-datasaurus-dozen.html\n",
"- https://thenewstatistics.com/itns/2017/05/11/what-the-datasaurus-tells-us-data-pictures-are-cool/\n",
"- https://www.autodesk.com/research/publications/same-stats-different-graphs\n",
"- https://damassets.autodesk.net/content/dam/autodesk/research/publications-assets/pdf/same-stats-different-graphs.pdf"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "Python 3.8.12 ('base')",
"language": "python",
"name": "python3"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.12"
},
"orig_nbformat": 4,
"vscode": {
"interpreter": {
"hash": "fc07faea01ee7d3acb1f2aba12b38e3e091e74ac7d919650e214d6284dcb7e46"
}
}
},
"nbformat": 4,
"nbformat_minor": 2
}
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment