Tools of the Trade #1: Plot

Wheat250-1

In an earlier post, I proposed to differentiate naïve from skilled statisticians, and design graphs accordingly. The tool that I recommend here is very useful to display simple matters: it produces excellent graphs for naïve statisticians. It is a small, specialized, free program for Apple’s Mac Os X, and it’s called – Plot.

Just like many other mac applications, plot benefits from simplicity. It draws two-dimensional graphs: line charts, data points or histograms, and – well, that’s it. It takes data from a variety of basic file formats or – a nice feature – pull it from sql databases. What sets plot aside from more widespread solutions, like Excel, SPSS, R, Statistica and others, is its ease of use and the quality of its output (cf. example in fig.1). Once a file is loaded, line styles can be adjusted for each variable. All that is left to do is placing the legend (maybe removing graphical overhead) – et voilá, a beautiful, minimalist graph has been created in minutes.

Plot-Ex-2

The great advantage is that plot strikes a balance between functionality and usability: it is not as awkward to use and its output is a lot better than Excel’s, while at the same time the user interface (fig.3) stays simple. Its reductionist approach seems to be beneficial in another way: instead of trying out more and more complex visualizations (“Oh wait, maybe a 3D scatterplot could…”), the user decides on the type of chart first – and by choosing plot it will automatically be a type naïve statisticians understand.

Verdict: If I can visualize data with plot, the resulting graphs will look elegant yet precise, without superfluous distracting elements (boxes, large symbols, color background &c.) – and because of its form, I will be able to explain it to anyone without first handing him/her a statistics manual.

Plot-Ui

References:

The Plot Homepage – http://plot.micw.eu/

Advertisements

About this entry