When I really realized that I love the visualization of data even more than just the analytics of data, I went around searching for information on how you might become a data visualization designer/specialist. Especially if you do not have a degree in a field such as design. I’ve gathered resources on book, tutorials and blogs which I wanted to share with you.
Icon explanation ♦ Added bonus if you're a d3.js user ∴ You might need a subscription to view all content / courses
The book section was getting too big so I put them on a separate page. On this page you can find my reviews and thoughts on the content and usefulness of books that I’ve read. I hope this will assist you to figure out which books you still want to read to learn more about data visualization.
Inspiration from Pinterest
Although there is no text on Pinterest to guide you to become a data visualization designer, I, and many others, use it as a treasure trove of inspiration, which is very important (multiple pages tell you to “Steal like an artist”). Follow boards that inspire you and collect those pieces that resonate best in your own boards. Some names to get you started
Articles about learning data visualization
It can’t really get any more relevant than this. It outlines the ideas to get started, gives a good starting book list with explanations on why you should read them (and which influenced my book list below) and tells you some straight answers: “You have to learn how to code” & “Visualization design is something you cannot learn if you don’t practice”.
Also from the FILWD blog by Enrico Bertini. This post explains what data visualization tools there are and how you can choose the tool that is right for you (#1 is actually in the book list).
Especially one of the earliest podcasts with Andy Kirk “How To Learn Data Visualization”. Almost every episode revolves around a specialist in the field of data visualization or there is a general subject such as color. The extra links to material discussed for each episode are also very useful. Each podcast takes about an hour and they already have more than fifty so, talking from experience, it can take a while to catch up, but 95% of the episodes are worth it.
I found this to be an extremely good piece with the most valuable tips to create good data visualizations (that are not listed on typical “step” lists too often). I’ve read books that talk about these steps, but never one that combines all of these fields (steps based on psychology, on business, on common sense).
This page of Jeffrey Heer’s course on Data Visualization offers a good overview of freely available research papers on visualization.
A free online e-book about how to become a Data Journalist. This book has multiple case studies of the best data journalism departments around the World (practically always including data visualizations, there is even a section called “Visualization as the Workhorse of Data Journalism”), but it also has sections on getting, understanding and delivering data. Perhaps not a lot about how to create an actual visualization, but you need so much more than just the visual to make something engaging and that’s where this book really helps.
The best parts of the AMA with Mike Bostock about learning how to do data visualization (in d3.js) have been combined in this nice post.
This is an enormous collection of the slide decks from presentations and tutorials that were presented at NICAR, a yearly conference devoted to data journalism. Topic range from introductions to R, d3, Gephi, more general data wrangling and lots more. There are also links here to the decks from previous years.
Interesting article about how there is only one good answer to questions about the design and presentation of a chart: “It depends”.
How can anyone start a career in data visualization and […]? & How can one carve out a career in data visualization + graphic design?
Actually two questions on Quora but with a few extensive answers by world renowned data visualization practitioners.
Very kind that Andy Kirk does all of this hard work to collect really good pieces of data visualization. Not just visualizations but also articles, news and, very important for this section, learning & development. The link above points to his most recent collection at the time of writing (Feb 2015), but to see each and every month (he started in Feb 2010 I think) just visit his blog and select the collections.
It shouldn’t be needed, but I do include it to show you that it’s a nice long list already, so hopefully many more female data visualizers will join.
Converting the tips from William Zinsser’s “On Writing Well” and applying this to data visualization.
Articles about how experts are designing their work
I cannot say often enough how much I love the work of Giorgia Lupi and her team at Accurat! In this post, with many nice sketches, she outlines her technique of going from an assignment to a design sketch that can be used for the execution on the computer.
Another, more recent piece about the design process at Accurat, more specifically about their amazing “La Lettura” works. With many great examples.
A post by Fernanda Viégas & Martin Wattenberg. Since design is not a science there are no clear cut answers to what might be the best way to visualize a dataset. This post explains why critique by peers is a good thing to get to a higher level. Luckily, they also explain how to give a useful critique.
A nice blog on the creation of the more intricate science driven infographics with several really good links to other great content.
Another one on the Scientific American. This blog is looking into the entire design process of one specific infographic that was made for the publication “Secret Life in Household Dust”.
An article written by one of the best creators of data journalism (in my opinion) about how and why he is taking the approach to data journalism that sets him so apart from many others.
A very long, but packed with useful tips, post on the creation of a beautiful (d3 based) data visualization. It really shows the process of how they came to the end design. From different layouts, how to handle large datasets in the browser to choosing a color palette. All illustrated with many examples.
A gorgeous data visualization and actually build with NodeBox. After reading this I was truly intrigued by NodeBox (but I only have so much free time). This blog outlines the entire design process, from first tries to mistakes, to it slowly evolving into the final poster.
In this guest post on Visualoop.com Sonja explains how she created the gorgeous Weather Eindhoven 2014 charts from idea to final product. Perhaps a bit short, but still interesting.
Another post on Visualoop (and again about weather) explaining the process of going from idea to data to final design.
An awesome article from the Wall Street Journal about the lyrics of the Hamilton Musical. I always like to read that such great stuff doesn’t just come from nowhere, that they went through many iterations before ending up in the final wonderful (elegant & simpel) result.
An interview with David McCandless by Vox. He has a unique style that you easily recognize once you’ve seen a few of his works. On a high level he explains the process he goes through from start to a finished piece of art. Some months afterwards he created this nice Venn diagram on “What makes a Good Visualization”.
Articles about useful best practices to know
Infoposters are not Infographics & Random thoughts on infographics, simplification, and the revolting 10 seconds rule
Two post that explain the difference between what a true infographic entails and what most people (sadly) now think is an infographic.
A blog that focuses on typical mistakes that a designer can make during a client presentation. Some of it is not 100% spot-on for data visualization, but there are enough good learnings in here to make it a worthwhile read.
These are Kennedy Elliott’s speaker notes from the talk she gave at OpenVis 2016. Very useful to have so many studies summarized and bundled in one post (including many images explaining the key takeaways).
Storybench often has very interesting articles on data visualization. This article is a write-up of an interview with Michelle Borkin who recently published a study titled “Beyond Memorability: Visualization Recognition and Recall”. Read about the key points (and hopefully remember them for your next dataviz).
An important part of data visualization is the data of course. And in the real world data is often perfect, missing or plain incorrect values. This guide will give you tips on how to resolve many of the most common issues.
Often a question and often wrongly applied by those starting with data visualization (there’s no definitive answer, but there are definitely cases where it should start at zero. Check out the link to the Vox video at the bottom for a great explanation why it isn’t always needed.
Another online d3 page. In terms of content I would say it covers the same as the one above, so choose the site which you think works best for you.
Alberto Cairo & Scott Murray have given several Data Visualization and d3.js courses in the past few months. Who wouldn’t love to get a course from such great teachers (definitely check out their books as well).
This is a 7 week online course about Data Visualization. It covers how to apply design principles, human perception, color theory, and effective storytelling to data visualization. I haven’t taken it, so I cannot comment on how good it is, but from the website I can see that more than 25k people have already started it. O, and it appears to be free.
Perhaps you prefer a format where you can sit back and watch how somebody else shows you how to do it. Then this 1.5 hour tutorial will just be the thing for you.
The 4th version of d3 has had quite some changes and updates. Irene Ros created a deck outlining practically all of these changes and what unexpected new things you’re able to do.
All access to Lynn Cherny’s class on interactive data visualization; 15 weeks of amazingly rich content & resources, elaborate explanations with code. A really great resource to learn d3.
A list of d3.js tutorials written by Mike and many others. Also has a summary of d3.js books and other media from which you can learn about d3.js.
d3.js basics ♦
Although some are present on the tutorial list above, since margins, updates, selections and nests are such a key element to d3.js, no matter what you build, I wanted to list these tutorials separately (also note that some are based on d3 v3 and other on v4, and other again are version independent)
- Margin Convention
- How Selections Work
- Practical applications of a d3.js selection
- Nested Selections
- Manipulating data like a boss with d3
- Thinking with Joins
- Understanding selectAll, data, enter & append
- General Update Pattern I
- General Update Pattern II
- General Update Pattern III
- Object Constancy
- Nesting CSV data
- Mister Nester - Experimenting with nest
- Responding to resize
- Improving d3 Path Animation
d3.js options ♦
Some of the d3 settings have many options and it can be difficult to remember what each of these options might do. The sites below are places I keep coming back to to understand what option I am looking for this time
- Visual effects of interpolation functions
- Visual effects of the 9 easing functions (on transitions)
- Easing in d3v4 is been expanded quite a bit. In this block you can see how each of these looks and feels
- d3 time formatting examples to see what the format returns
Several very nice and elaborate easy design choices that can make all the difference, not exclusively for UI design. Written by an engineer so it comes from the right mindset for me.
There’s even a part 2 that goes more into the use of text and fonts.
A blog showing how using the right kind of animation principles can make a user interface more friendly and fun to use (but imagine how it might apply to more complex data visualizations). For example, I incorporated the Gooey effect that can be seen on this blog in a short blog.
An extensive list of mostly video tutorials that teach the essentials of Adobe Illustrator. I’ve only seen the first 15-20 video’s but I intend to watch them all at some point.
I think this compares best with the more broadly known Lynda.com in that it offers (short-ish) online courses made by experts in the field and those fields are mostly related to design, typography and photography, although Skillshare might focus even more on design than Lynda. There is also a technology section which deals with web mostly. It’s not free (only $8 a month though), however if you subscribe to their newsletter you get a weekly digest which also includes several classes that are free to enroll for a week (once you enroll, you can take the class whenever you want). I’ve done a few free classes and really like the quality and diversity.
I specifically wanted to point out this course by Nicholas Felton on Skillshare. I haven’t done it but it would be the first one on my list once I take a subscription.
Similar to Skillshare but around for a longer time I think. It’s more expensive, starting at $25 a month for the most basic subscription. I haven’t actually taken any courses here yet, only made a list of courses that I’d like to do if I ever decide to try it out for a month. I’m sure the quality is good, but for now I get a better feeling from Skillshare than Lynda.
TED talks are some of the best talks that are out there (just watch those by Hans Rosling) and this is a nice list with 20 talks that cover topics such as: how to think more creatively, finding happiness in work or learning something new.
Introduction to R ∴
It’s been too long since I took any R courses to say what is good nowadays. In general I would advice something that lets you code alongside as you learn. Maybe something in R, maybe something in the browser itself (that still uses R of course). Something that goes into the
tidyverse is probably good. The best way to learn a new programming language is by doing it!
Although not a real R tutorial, but you’re probably going to come across some for of text analysis (or dirty data that needs cleaning) and then you’re life is going to be much easier of you regex. This site makes it very easy to learn and test your regex expressions before applying it to you dataset.
An extensive explanation of the creation of the chart provided in Edward Tufte’s classic book Visual Display of Quantitative Information, 2nd Ed. (page 30) using the ggplot2 package that makes you think again about R not being able to create beautiful and intricate charts.
For the creation of a static image during one of my work analytics projects which also need to end up in the final presentation for the client, I always use ggplot2. It might take some time to get the hang of the coding, but once you do, you realize its actually very logical and easy.
Halfway on this site you get to a section with all kinds of links to different subjects in R. These lead to very well documented PDFs that explain how to do it (with the code) and what the output of R should be. It starts with simple things such as reading data into R, but gets more advanced towards the machine learning techniques.
A very complete tutorial to create data visualization according to Tufte’s standards. They won’t be very eye catching, but they’ll be straightforward, truthful and packed with data.
I’ve done many MOOCs on Coursera to learn more about Data Science. Many of the data analytics courses use R and some are real introductions into the domain of data collection and data cleaning which are also useful for any data visualization designer who is doing data analysis and cleaning as well (which I hope is many of you).
Another very extensive tutorial on creating a nice data visualization completely in R about the change in colors of paintings over time. The code is not for a beginner, but I added it nonetheless for those more experienced in R (or as a nice project to try for yourself once you finished all the tutorials above).
HTML widgets work just like R plots except they produce interactive web visualizations. A line or two of R code is all it takes to produce a d3 graphic or Leaflet map. HTML widgets can be used at the R console as well as embedded in R Markdown reports and Shiny web applications. Here you can find good examples of output and code.
In data visualization you’re often going to come across JSON files with your data. So a tutorial on how to pull JSON’s into R and then do some data preparation is very handy.
A package that turns your R graphics into light-weight SVGs (instead of the extremely high accuracy and thus large files that R’s svg engine gives you).
Same place as above but this time about HTML & CSS. I also still have to take this course, but it looks like a good one.
Since d3.js builds with SVG I believe it wouldn’t be bad to know a bit more about SVGs in general.
This is a free section of the larger HTML and CSS course and it dives into the basics of HTML which you will definitely need to know when you want to place your data visualization in a nice looking website.
I know that was a very long list, but hopefully by providing you with lots of options you can select exactly those links that will fit your style. Wether or not you can or want to code or prefer to stay with the more graphic tools like Illustrator. Good luck!