Communicating data clearly

By Grant Jacobs 25/10/2012

Correct and appropriate presentation of graphs matters. It’s worth taking time to consider how to present your data, to convey the information clearly in a way that is readily perceived and accurate.

Not just for scientists, either. Graphs are used ubiquitously, after all.

‘Convincing’, xkcd. Original:

On-line there is some excellent material on presenting data. One example is a handout for a presentation[1] Communicating data clearly by Naomi Robbins, who writes Effective Graphs at Forbes blogs. (Some of us here take down poor science coverage in the media. Among other things Naomi writes, she takes down poor graphs in the media!)

Many forms of presentation fail to convey the data well; some are simply confusing. This handout is limited to key points, but I hope that it might encourage readers to think about their data presentation more carefully.

(From handout, clipped to suit blog post format.)

Those familiar with the topic will know that part of the issue is how we perceive visual elements such as angles and areas. In this way presentation of graphical information intersects with cognitive neuroscience – another example of practical applications from what might otherwise be perceived as ‘blue skies’ science. As you might expect, Naomi’s handout includes some of these aspects, too.


Another short post while I continue to tackle the tax return.

1. As part of NYC data week. Computational biologists might recognise this as including O’Reilly Media’s Strata + Hadoop World Conference.

Other articles on Code for life:

On vetting TED(x) events – a suggestion

Thoughts on scientific abstracts also a science writing check-list

One example of why all those genomes from different species are useful to biologists

Fast access to biological databases from your web browser

Reproducible research and computational biology

0 Responses to “Communicating data clearly”

  • Another good resource (though it requires a subscription) is the Points of View series by Bang Wong in Nature Methods. He’s covered a lot of topics from color schemes to multidimensional data viz.