The Art of Statistics: Code, Data, Errata and Additions

UK hardback: UK paperback: US hardback:

Here is the book’s UK Amazon page and US Amazon page

The UK and US hardback versions are identical in content. Errata and additions are listed below for all versions.

This file is produced by R Markdown, and there is also a web-page produced by Github.

Code repository

The repository contains data and R code for the Figures and Tables in Art of Statistics. They should all work in RStudio.

It is not yet complete, and will change and improve.
The graphics were originally mainly produced in basic R, but (with much appreciated assistance) these have been mainly revised into ggplot2 - sometimes both versions are provided for comparison. However I do not pretend to have any particular skill in using R or ggplot, and no doubt many improvements could be made.
Many packages are used and will need to be installed, for example using install.packages(“ggplot2”)
While the individual files can be downloaded, it is probably easiest to download a zipped repository using the green download button.

Suggestions for improvements to david@statslab.cam.ac.uk

Markdown output for individual Figures

Errata and additions - Paperback

Many apologies that errors are still there (sob).

page xiv list of tables Delete the entry "12.1 Questionable Interpretation and Communication Practices 354"
page 146 para 3, line 4, replace "dislikes" by "reactions"
page 150 legend to Figure 6.1 replace "to a Francis William Somerton" by "to a William Henry Somerton"
page 172 line -1. replace 'selects relevant' by 'eliminates unnecessary'
page 179 para 2, line 5 replace "When a vision algorithm was trained to discriminate pictures of huskies from German Shepherds, it was very effective until it failed on huskies that were kept as pets – it turned out that its apparent skill was based on identifying snow in the background" by "A vision algorithm intended to discriminate pictures of huskies from wolves was (deliberately) trained on images of wolves in snow and huskies without snow, with the result that any future image of either dog was classified as a wolf if there was snow in the background"
page 182 line 12 “Communicated” -> “communicated” (small “c”)
page 183 Table 6.5, line 5 replace "Trastuzamab" by "Trastuzumab"
page 191 line -1 it should read "there were 1,215"..
page 221 line 10 “quantity” -> “number”
page 223 line 13 “driven” -> “that can be modelled”
page 249 para 2, line -2 Put single quotes around "confident"
page 265 line 3, replace "null distribution" by "distribution of the observed difference, were the null hypothesis true"
page 272 legend to Table 10.3, line 2 replace '2014' by '2013'
page 286 line 6. Add ‘1 -’ in front of $\beta$
page 286 line -7. Add ‘1 -’ in front of $\beta$
page 322 lines -6, -8, -11 Replace "royal" by "straight" in three places
page 346 line -11 before "two groups", add "the changes in"
page 385 Glossary entry for chi-squared tests. Add at end 'For small counts a 'continuity correction' can be applied, and this is used for the data in Table 10.2'
page 396 Pearson Correlation Coefficient. The formula is missing a square root sign in the denominator, and should read
$r = \frac{ \sum_{i=1}^n (x_i - \overline{x})(y_i - \overline{y}) } { \sqrt{\sum_{i=1}^n (x_i - \overline{x})^2 \sum_{i=1}^n (y_i - \overline{y})^2 }}.$
page 396 Pearson Correlation Coefficient. Add $\frac{1}{n}$ before $\sum_{i=1}^n u_i v_i$
page 396 Pearson Correlation Coefficient. Add after ‘Z-scores’ ‘(this assumes the standard deviations have been calculated with $n$ in the denominator: if $n-1$ has been used, the formula is $\frac{1}{n-1}\sum_{i=1}^n u_i v_i$
page 409 note 5 replace "https://esa.un.org/unpd/wpp/Download/Standard/Population/" by "https://population.un.org/wpp/Download/Standard/Population/"
page 413 note 4 replace with "https://www.kdd.org/kdd2016/papers/files/rfp0573-ribeiroA.pdf"

Errata and additions - Hardbacks

(these should now all be corrected in the paperback)

page 26 Figure 1.1 At the end of the legend, add ‘Rather than a bar-chart, it may be better to use dots for the data-points when the axis does not start at zero.’
page 49 Footnote. ‘31,337’ was probably a deliberate choice by an ageing geek, as it was an old expression for ‘eleet / elite’
page 59 Legend for Figure 2.6, replace ‘Alberto Cairo’s’ by ‘the’
page 71 line 2, add ‘before having their first child’
page 86 line -9. ‘2,190’ should read ‘2,910’
page 110 The ‘M’ in STEM should be ‘Mathematics’ and not ‘Medicine’
page 156 line 9. ‘large’ should be ‘smaller’
page 159, 161 The question on page 161 should be before the last two lines on page 159.
pages 191 to 200 - see below
page 206, header says ‘Chapter 7’, should be ‘Chapter 8’
page 212 Figure 8.3 – the label ‘Head’ on the first lower branch should be ‘Tail’
page 222, line -3, ‘2014’ should be ‘2013’
page 225 legend to Figure 8.5, ‘2014’ should be ‘2013’
page 232. Figure 9.1b is incorrect - the bars should have heights 0.32, 0.64, 0.04, as described in the footnote. Figures 9.1e and 9.1

ArtofStatistics

Install / Use

README