Before-After/Control-Impact (BACI) Experimental Design

Last Updated: Sept 12, 2012

Challenge: Visualize the magnitude of impact (a.k.a experimental effect) by showing all of the observations, rather than just the results of a statistical test. Also, set up the comparison to check whether the underlying assumptions of the statistical test are met (e.g. should the sample be stratified along some background variable, such as age, depth, or position along a transect).

Solution: Side-by-side bar charts to compare the two samples; visually connect the means of the two panels; flip back-and-forth between two pages to show the difference between before and after

StackedBars_thumb

Developed for: FigRs, to illustrate different uses of three simple R graphing tricks