Charts and infographics

One of my supervisors, Sama Low Choy, is always up for a chat about the role of visualisation in statistics and data analysis. An acolyte of Tufte, Sama is quite passionate about appropriate methods of presenting data. 3D pie charts, needless to say, are public enemy number one.

We were having a chat this afternoon about how a friend of mine who’s about to start a Masters in biology/statistics is a bit of a nerd when it comes to the presentation of data and copy. Talk turned to how easy it is to use LaTeX and how there’s a need for better visualisation in scientific results (something I’ve talked to some of my ILAQH colleagues about at length). Infographics are quite a popular thing at the moment, despite not always being particularly informative nor warranted, and I mentioned a blog about the process that the New York Times crew go through when developing infographics and other visualisation tools.

Sama has forwarded me an email containing some of her favourite links that deal with the field of “data journalism”.

The Times in US provide some inspiration:

Even on a topic that some science & engineerings students may become engaged with:

Someone else’s selection of highlights from their Infographics

Here is a link to a website about a new (free) handbook on data journalism:

Here’s a link to some videos (showing some software, doing some intuitive roll-up roll-down exploration & aggregation of data):

You might also enjoy these videos of Amanda Cox, a woman behind some of the more innovative visualisation pieces at the NYT, talking about the processes of making good quality visualisations.


6 thoughts on “Charts and infographics

    1. Sam Clifford Post author

      I thought you might. If you’re going to use R it might be worth learning ggplot2 as it seems to be quite good. The fellows over at Statisfaction seem to be quite fond of it. I find it difficult to use, preferring the base graphics tools that come with R.

      1. Luisa

        R is probably it for me, since it seems further in line with the mathemagical programming I’ve already done than any of the commercial stats packages. I will probably be asking you for advice on how to picture argh further down the track.

      2. Sam Clifford Post author

        Another option is to generate your results in R and then use pgfplots in LaTeX to generate your graphs. It’s much easier to use than straight TikZ.

      3. ihrhove

        You can always use tikzDevice for directly converting R graphics (even ggplot!) into tikz-code for latex. of course it gets messy and not all graphs work, but in 95% of the cases its a really good choice.

      4. Sam Clifford Post author

        I had a look at TikZDevice maybe two years ago and it looked promising then but it still did give a lot of weird stuff. I might have to give it a go again. I really tried getting into using pgfSweave but ended up getting stuck on some things so gave up. I should at least give pgfplots another go.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s