Quick Visualization Toolbox

This is a set of custom functions for plotting various data visualizations on Matlab. If you want to try more in-depth vizualisation, I strongly recommend Gramm, which is fantastic for that purpose. In contrast, the purpose of this repository is to provide a light, easy-to-use, and self-contained toolbox for quickly producing publication-friendly figures.

The optional argument syntax is identical across all functions, allowing fast and seamless switch between different functions types and a smooth familiarization phase for new users.

Run example_plots.m to reproduce the example figures below.

Plotting distributions

Gauge plots for categorical distributions

The function gaugeplot.m produces bar plots with divisions proportional to categorical groupings.

Distribution plots for uni-dimensional continuous distributions

The function distline.m produces kernel-smoothed distributions on a one-dimensional axis. The properties of the kernel can easily be adjusted using input arguments to the function.

Distribution plots for two-dimensional continuous distributions

The function scatterdist.m produces scatter plots with projected distributions on each axis. The Distributions can be "classic" histograms or kernel-smoothed distributions. The properties of the kernel can easily be adjusted using input arguments to the function.

Plotting figure legends

The functions legendbox.m and legendline.m allow plotting legends anywhere on a figure. An illustration of the legend produced by legendbox.m and legendline.m is available on the right-hand side of the gaugeplot and distribution plot figures above, respectively.

Plotting group data

For all of the functions below, the mode and error interval can be defined by the user in the same way. Users can chose between mean and median for the mode, and between standard deviation, standard error of the mean, and bootstraped confidence interval of the chosen mode.

For all functions below, individual datapoints can be plotted with a user-defined layout:

• 'align' aligns all datapoints on an invisible line.
• 'deport' aligns all datapoints based on the data's cumulative distribution.
• 'jitter' jitters all datapoints according to their kernel-smoothed distribution. The properties of the kernel can easily be adjusted using input arguments to the function.
• 'silo' jitters all datapoints randomly in a "silo-like" column.
• 'none' removes individual datapoints entirely.

You can see an illustration of each of these options in the figures below, in the same order (from left to right).

Bar plots

The function barreplot.m produces typical barplots as used in most publications.

Dot plots

The function dotplot.m produces barplots without the bars, just the dots.

Boite plots

The function boiteplot.m produces typical boxplots as used in most publications. The mode and error interval can also be added alongside the quantile-providing boxplot.

Rain plots

The function rainplot.m produces rainplots, also sometimes called half-violin plots. The "cloud" is produced in a similar way as distlines.m, using a kernel-smoothing distribution. The properties of the kernel can easily be adjusted using input arguments to the function.