data-visualisation/datasaurus.Rmd at master · deanmarchiori/data-visualisation · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
---
title: "datasaurus"
author: "Dean Marchiori"
date: "12/31/2020"
output: html_document
editor_options:
  chunk_output_type: console
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```


# 1 min
```{r}
library(tidyverse)

datasaurus <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-10-13/datasaurus.csv')

ggplot(datasaurus, aes(x, y)) +
  geom_point() +
  facet_wrap(~dataset)
```


# 5 min
```{r}
library(tidyverse)

datasaurus <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-10-13/datasaurus.csv') %>%
  filter(dataset != 'away')

ggplot(datasaurus, aes(x, y, colour = dataset), alpha = 0.8) +
  geom_point(pch = 21, size = 1, show.legend = FALSE) +
  facet_wrap(~dataset) +
  theme_minimal() +
  labs(title = "Beware the Datasaurus Dozen",
       subtitle = "Each dataset has five statistics that are (almost) the same in each case, but look very different",
       x = "",
       y = "")
```

# 20 min
```{r}
library(tidyverse)
library(purrr)
library(broom)

datasaurus <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-10-13/datasaurus.csv') %>%
  filter(dataset != "x_shape")

x <- datasaurus %>%
  group_by(dataset) %>%
  nest() %>%
  mutate(models = map(data, ~lm(y ~ x, data = .x))) %>%
  mutate(clean = map(models, broom::tidy)) %>%
  unnest(clean) %>%
  select(dataset, term, estimate) %>%
  pivot_wider(names_from = term, values_from = estimate)

ggplot(datasaurus) +
  geom_point(aes(x, y, colour = dataset), show.legend = FALSE) +
  facet_wrap(~dataset) +
  geom_abline(data = x, aes(slope = x, intercept = `(Intercept)`), colour = "red") +
  theme_minimal() +
  labs(title = "Datasaurus Dozen",
       subtitle = "Showing fitted linear regression line. While each dataset looks different, each has five statistics that are (almost) the same in each case")
```