The Stack Overflow podcast is back! Listen to an interview with our new CEO.

The design of experiments (DOE) is the study of how to structure an information-gathering exercise where variation is present. In statistics, controlled experiments are usually implied, meaning that study units are randomly assigned to conditions where levels of a variable have been independently manipulated (prototypically into treatment & control groups).

The design of experiments (DOE) is the study of how to structure an information-gathering exercise where the relationship between variables is not deterministic (i.e., at least some random variation is present). In statistics, controlled experiments are usually implied, meaning that study units are randomly assigned to conditions in which the level of a variable[s] has been independently manipulated (minimally, and prototypically, to create a treatment condition and a control condition).

In the design of experiments, the experimenter is often interested in the effect of some process or intervention (the "treatment") on a response variable for some objects (the "experimental units"). True experiments are typically thought necessary for inferring causality, although there is a field of study within statistics that concerns itself with how causality may be nonetheless inferred from observational data.

history | excerpt history