What do you think?
Rate this book
Data Visualization: A Practical Introduction
An accessible primer on how to create effective graphics from dataThis book provides students and researchers a hands-on introduction to the principles and practice of data visualization. It explains what makes some graphs succeed while others fail, how to make high-quality figures from data using powerful and reproducible methods, and how to think about data visualization in an honest and effective way.Data Visualization builds the reader’s expertise in ggplot2, a versatile visualization library for the R programming language. Through a series of worked examples, this accessible primer then demonstrates how to create plots piece by piece, beginning with summaries of single variables and moving on to more complex graphics. Topics include plotting continuous and categorical variables; layering information on graphics; producing effective “small multiple” plots; grouping, summarizing, and transforming data for plotting; creating maps; working with the output of statistical models; and refining plots to make them more comprehensible.Effective graphics are essential to communicating ideas and a great way to better understand data. This book provides the practical skills students and practitioners need to visualize quantitative data and get the most out of their research findings.
Provides hands-on instruction using R and ggplot2Shows how the “tidyverse” of data analysis tools makes working with R easier and more consistentIncludes a library of data sets, code, and functions
Provides hands-on instruction using R and ggplot2Shows how the “tidyverse” of data analysis tools makes working with R easier and more consistentIncludes a library of data sets, code, and functions
296 pages, Kindle Edition
First published December 18, 2018
34 people are currently reading
671 people want to read
Ratings & Reviews
Friends & Following
Create a free account to discover what your friends think of this book!
Community Reviews
Displaying 1 - 20 of 20 reviews
January 3, 2019
In the preface to the Data Visualization: A Practical Introduction author Kieran Healy writes:
The book itself is also beautifully designed. The use of figures and margin notes give you a sense of being guided through the ideas rather than just being told what they are. I've had lots of fun going back to some of my own visualizations made with R and ggplot2 and improving them based on what I learned here.
I absolutely recommend this to beginners and experts alike. Healy gives you everything you'd need to know if you're starting from scratch, but in such a way as to not slow things down for the more experienced reader. For that reason, it would also make a great book for a course on applied use of R.
My main goal is to introduce you to both the ideas and the methods of data visualization in a sensible, comprehensible, reproducible way.Well, mission accomplished. The book is at once enormously readable, and sufficiently technically detailed as to make it easy to implement the principles introduced.
The book itself is also beautifully designed. The use of figures and margin notes give you a sense of being guided through the ideas rather than just being told what they are. I've had lots of fun going back to some of my own visualizations made with R and ggplot2 and improving them based on what I learned here.
I absolutely recommend this to beginners and experts alike. Healy gives you everything you'd need to know if you're starting from scratch, but in such a way as to not slow things down for the more experienced reader. For that reason, it would also make a great book for a course on applied use of R.
April 9, 2021
This actually is a good book for learning how to visualize data. I learned a lot from what I read, but in the end, it basically taught me that my best bet is to find collaborators for whom learning the language of "R" comes more naturally. For someone more adept at learning these techniques, this book would likely be much more helpful than it ended up being for me. Although I did learn some basics and could see myself referring to it in the future if I have some other basic tricks I need to relearn.
November 12, 2019
An exceptionally good introduction to data visualization using R mainly focusing on the tidyverse and ggplot packages along with Healy's own socviz package. The author doesn't just show how to construct sensible, visually appealing plots he explains why you should consider doing so.
There is a lengthy discussion at the beginning about elements of design and perception, as well as a short discussion on the R language, tidy data, and ggplot geoms. Soon, you are creating visualizations by incorporating maps of the US and understanding the limitations of such plots.
Overall, Data Visualization is organized very well with many thought-provoking and informative examples. If you are a researcher who needs some inspiration for how to visualize your data, this book certainly provides an opportunity to discover some plots that you may not have thought of yourself.
There is a lengthy discussion at the beginning about elements of design and perception, as well as a short discussion on the R language, tidy data, and ggplot geoms. Soon, you are creating visualizations by incorporating maps of the US and understanding the limitations of such plots.
Overall, Data Visualization is organized very well with many thought-provoking and informative examples. If you are a researcher who needs some inspiration for how to visualize your data, this book certainly provides an opportunity to discover some plots that you may not have thought of yourself.
August 28, 2019
What if I told you that you would never have to make a chart in Excel ever again? Never again having to trick it into accepting your axis settings, never getting exactly the wrong kind of line graph every single time? (Seriously, how does it get it wrong so consistently?)
Well for the low price of having to learn a little R, you could live this dream. If that sounds scary, don't despair, because Kieran Healy's Data Visualization: A Practical Introduction will walk you through everything ggplot-related. I had some R experience to start off with (and I would recommend at least a minimum of coding/data work as a background), but I think this book is fine for R newbies. It uses a package created for the book, so you can follow along with all examples in the text, and you should. I didn't think state-by-state or county-by-county map plots were something I could do, but thankfully it's not as hard as I thought it was - although there are some unique considerations (like projections) I didn't know I had to think about. The book isn't perfect and some examples didn't work for me, likely due to changes in packages, but overall I found it incredibly useful.
Plus it’s entirely online at https://socviz.co/! Beautiful visualizations are within your reach! Just keep calm and add geoms.
Well for the low price of having to learn a little R, you could live this dream. If that sounds scary, don't despair, because Kieran Healy's Data Visualization: A Practical Introduction will walk you through everything ggplot-related. I had some R experience to start off with (and I would recommend at least a minimum of coding/data work as a background), but I think this book is fine for R newbies. It uses a package created for the book, so you can follow along with all examples in the text, and you should. I didn't think state-by-state or county-by-county map plots were something I could do, but thankfully it's not as hard as I thought it was - although there are some unique considerations (like projections) I didn't know I had to think about. The book isn't perfect and some examples didn't work for me, likely due to changes in packages, but overall I found it incredibly useful.
Plus it’s entirely online at https://socviz.co/! Beautiful visualizations are within your reach! Just keep calm and add geoms.
November 21, 2022
I read the incomplete draft the author posted free online (https://socviz.co), and I don't know the difference between this draft version and the published version. But even from this draft, I can say this is a great book. I recommend it to everyone who uses ggplot2 or its implementation in other languages (e.g., plotnine in Python, my plotting solution). Even if you use other tools instead of ggplot2, you can still learn from the book good practice of visualization through easy-to-reproduce examples.
The author includes the basics of ggplot2 to make the book accessible to beginners. So, depending on your familiarity with the package, feel free to skip some chapters or sections. But an experienced user can still benefit from reading this book. For example, I've used ggplot2 for a few years, but I didn't know each geom layer can use a different dataset, allowing one to highlight or label only part of the graph. I also felt more confident in using text labels and annotations in charts after knowing they are not as difficult as I thought.
I use ggplot2/plotnine because the underlying grammar of graphics is more consistent and easier to memorize than those of other tools (base R, lattice, matplotlib, seaborn, SAS, Matlab). But even so, I was still confused by some of its syntax (why set stat = 'identity'? what is the ..prop..? Why sometimes I need to set 'group', and sometimes I don't?) The author's discussion of what is under the hood of the syntax helped me solve these mysteries. For example, each geom_ function has an associated stat_ function that it uses by default, and the ..xxx.. are the stat variables temporarily computed by the stat_function. By default, geom_bar uses the ..count.. variable to draw the y axis, but if you set aes(y=..prop..), it will use proportion instead.
When the x variable in a graph is discrete and the y variable is continuous, the author suggests flipping the axes in the style of a Cleveland dot plot in the following situations: 1) when there are many discrete categories, which would make it too crowded to lay out all of them on the horizontal axis; 2) when the categorical labels are long and hard to read vertically; 3) when one wants to show the comparison of category frequency across faceted groups (See Figure 5.3 and 5.14 for examples). I find this recommendation reasonable and practical.
ggplot2 has a flourishing developer community (https://exts.ggplot2.tidyverse.org/), who invented many excellent extensions to ggplot2. The author mentioned a few in his book:
- ggthemes: make a graph in the style of the Economist or Wall Street Journal.
- ggrepel: create non-overlapping text labels to scatter points.
- ggridges: create partially overlapping density plots to show trend of distribution.
- viridis: create colorblind-robust palettes.
Even though I am not sure whether I can find the counterparts of these extensions in Python, I find it worthwhile to know the various ways people try to improve visualizations.
Side notes:
- Use Alt+minus in RStudio for Windows to type the assignment operator.
- The author provided a few caveats on interpretations based on maps. My summary: maps can be misleading because: 1) people tend to weigh the importance of each shaded/colored area by its size and ignore the (potentially vast) difference of population size and other characteristics between areas; 2) When the event of interest is not common, viewers' impression on the event per capita can be largely affected by rural areas with a low population size.
The author includes the basics of ggplot2 to make the book accessible to beginners. So, depending on your familiarity with the package, feel free to skip some chapters or sections. But an experienced user can still benefit from reading this book. For example, I've used ggplot2 for a few years, but I didn't know each geom layer can use a different dataset, allowing one to highlight or label only part of the graph. I also felt more confident in using text labels and annotations in charts after knowing they are not as difficult as I thought.
I use ggplot2/plotnine because the underlying grammar of graphics is more consistent and easier to memorize than those of other tools (base R, lattice, matplotlib, seaborn, SAS, Matlab). But even so, I was still confused by some of its syntax (why set stat = 'identity'? what is the ..prop..? Why sometimes I need to set 'group', and sometimes I don't?) The author's discussion of what is under the hood of the syntax helped me solve these mysteries. For example, each geom_ function has an associated stat_ function that it uses by default, and the ..xxx.. are the stat variables temporarily computed by the stat_function. By default, geom_bar uses the ..count.. variable to draw the y axis, but if you set aes(y=..prop..), it will use proportion instead.
When the x variable in a graph is discrete and the y variable is continuous, the author suggests flipping the axes in the style of a Cleveland dot plot in the following situations: 1) when there are many discrete categories, which would make it too crowded to lay out all of them on the horizontal axis; 2) when the categorical labels are long and hard to read vertically; 3) when one wants to show the comparison of category frequency across faceted groups (See Figure 5.3 and 5.14 for examples). I find this recommendation reasonable and practical.
ggplot2 has a flourishing developer community (https://exts.ggplot2.tidyverse.org/), who invented many excellent extensions to ggplot2. The author mentioned a few in his book:
- ggthemes: make a graph in the style of the Economist or Wall Street Journal.
- ggrepel: create non-overlapping text labels to scatter points.
- ggridges: create partially overlapping density plots to show trend of distribution.
- viridis: create colorblind-robust palettes.
Even though I am not sure whether I can find the counterparts of these extensions in Python, I find it worthwhile to know the various ways people try to improve visualizations.
Side notes:
- Use Alt+minus in RStudio for Windows to type the assignment operator.
- The author provided a few caveats on interpretations based on maps. My summary: maps can be misleading because: 1) people tend to weigh the importance of each shaded/colored area by its size and ignore the (potentially vast) difference of population size and other characteristics between areas; 2) When the event of interest is not common, viewers' impression on the event per capita can be largely affected by rural areas with a low population size.
This entire review has been hidden because of spoilers.
February 28, 2023
This truly is, as the subtitle says, a Practical Introduction. Kieran Healy writes about data visualization in R with a wonderful style, as if he is at your side teaching you (he is a Professor of Sociology).
Data Visualization is an emerging topic. What the title and subtitle are missing are the fact that this books relies heavily in the R programming language. So it was a bit hard to learn about data organization and visualization, at the same time that I had to understand R notation.
The book organization could use a bit rethinking. The Appendix, for instance, is very important and should be moved to an introduction - I can't undestand why the author waits for the end to explain why the plots you produce by following along the examples in the book do not match the plots in the book.
If you decide to learn R and data visualization, keep in mind that you have to look up a lot of details missing in the book. But this is also a good learning experience; in real life, data is badly documented.
Data Visualization is an emerging topic. What the title and subtitle are missing are the fact that this books relies heavily in the R programming language. So it was a bit hard to learn about data organization and visualization, at the same time that I had to understand R notation.
The book organization could use a bit rethinking. The Appendix, for instance, is very important and should be moved to an introduction - I can't undestand why the author waits for the end to explain why the plots you produce by following along the examples in the book do not match the plots in the book.
If you decide to learn R and data visualization, keep in mind that you have to look up a lot of details missing in the book. But this is also a good learning experience; in real life, data is badly documented.
July 24, 2025
L'idea del libro è molto buona, però mi aspettavo di più: in certi passaggi è fin troppo legato ad R, preoccupandosi di spiegare la sintassi e la struttura del codice (ottimo per i principianti, un po' noioso per gli utenti esperti).
Mi sarebbe piaciuto un discorso più concettuale, seppur concreto, su quale tipo di grafico utilizzare per quale tipo di dati, in funzione di ciò che si cerca di evidenziare, da aggiungersi ad indicazioni più esaustive sui colori (effettivamente presenti nel testo) e magari un compendio delle best practices.
La bibliografia offre numerosi spunti interessanti.
Mi sarebbe piaciuto un discorso più concettuale, seppur concreto, su quale tipo di grafico utilizzare per quale tipo di dati, in funzione di ciò che si cerca di evidenziare, da aggiungersi ad indicazioni più esaustive sui colori (effettivamente presenti nel testo) e magari un compendio delle best practices.
La bibliografia offre numerosi spunti interessanti.
December 16, 2020
Data Visualization by Kieran Healy is a fantastic ggplot() tutorial. Well designed, clear, concise, and right at my level: I have a solid foundation in basic R skills, know what I want, but needed an on-ramp on how to do it. This got me there. As others have pointed out, this book would also make for a great way to learn R. You'd start with data handling and visualization with this book, and then later move on to doing inferential statistics.
May 25, 2021
I liked it. The last few years Healy's blog has been very useful for data visualization, so I was happy to finally get my hands on a copy of this book. It's starting to be outdated just a tad, and there are some typos in the code, but nothing that someone with basic expeRience can't deal with. I'd recommend this book after R for data sciende. I picked up some new tricks and some other tricks I had forgotten about.
July 30, 2020
This is an excellent introduction to ggplot, the data visualization package for R. Of course, you could learn ggplot by googling but, in my opinion, having a reference book can be very helpful.
I liked that he respects the fundamental principles of data visualization. For example, he recommends using neither pies nor double y-axes, showing alternatives to both in ggplot.
I liked that he respects the fundamental principles of data visualization. For example, he recommends using neither pies nor double y-axes, showing alternatives to both in ggplot.
February 12, 2021
Data er den nye oljen? Hvis det er riktig så er datavisualisering måten man får oljen opp fra havbunnen. Datavisualisering gjør informasjon lettfattelig og er viktig for alle som jobber med trender og mønstre (investorer, markedsførere, kommunikatører og forskere, for å nevne noen). Denne boken gir en god innføring i denne praksisen.
Want to read
September 1, 2022"Along the way — as it turns out — I end up reminding myself of a lesson I’ve learned before about making sure you understand your measure before you think you understand what it is showing."
Daring Fireball: Visualizing Apple’s COVID Mobility Data
Daring Fireball: Visualizing Apple’s COVID Mobility Data
February 27, 2025
Great starting point for beginners, focusing on design principles and clear storytelling with data. It’s easy to follow, though advanced users might want more depth. Solid pick for clean, effective visuals!
July 15, 2018
I read a draft version of the manuscript. This is a very good book about data visualization for R users.
December 30, 2019
An outstanding introduction to data visualization. Clearly written and generously illustrated.
July 11, 2020
This is a great intro book for data visualization using R!! Prepare to ditch the old line and bar plots, get new ideas!
November 14, 2020
Incredibly useful. Following through this book greatly improved my skill in producing readable and aesthetically pleasing data visualizations. I recommend this to anyone working in R.
April 29, 2022
I've struggled to teach myself R for 2 years. This book is the most useful thing I've read. Both practical coding and has integrated visualization best practices. Love it.
October 9, 2020
A good practical introduction to the grammars of graphics
Read
March 4, 2019001.4226 H4348 2019
Displaying 1 - 20 of 20 reviews