Restricted randomization
In statistics, restricted randomization, also known as constrained randomization, occurs in the design of experiments and in particular in the context of randomized experiments and randomized controlled trials. Restricted randomization allows intuitively poor allocations of treatments to experimental units to be avoided, while retaining the theoretical benefits of randomization. For example, in a clinical trial of a new proposed treatment of obesity compared to a control, an experimenter would want to avoid outcomes of the randomization in which the new treatment was allocated only to the heaviest patients.
The goal of improving the balance between the treatment and control groups could be obtained using rerandomization, which is a form of restricted randomization in which allocations that fail a pre-specified covariate-balance criterion are discarded and the units are randomized again; this improves the precision of estimated treatment effects, but the analysis then has to account for the restriction. It has also been argued that the additional choices this involves, such as which covariates to balance and how to adjust p-values, give researchers more room for p-hacking. The concept was introduced by Frank Yates (1948) and William J.
Youden (1972) "as a way of avoiding bad spatial patterns of treatments in designed experiments."
Wikipedia contributors, "Restricted randomization", Wikipedia, revision 1382209490, retrieved 20 November 2026, source, CC BY-SA 4.0.