Unequal sample sizes according to the square‐root allocation rule are useful when comparing several treatments with a control

A common situation in experimental science involves comparing a number of treatment groups each with single reference (control group). For example, we might compare diameters fungal colonies subject to range inhibitory agents those from control group which no agent was applied. In this situation, the most commonly applied test is Dunnett's test, compares separately while controlling experiment-wise Type I error rate. analyses where all are treated equivalently, statistical power generally optimised by dividing subjects equally across groups. Researchers often still use balanced compared others. case, it fact optimal spread unequally: getting higher (n0) than k (n case). It has been previously suggested that simple rule thumb, so-called square-root allocation n0 = , offers better design, without necessarily being optimal. Here, show simple-to-apply substantial gains (over design) over broad circumstances and more-challenging-to-calculate design only minimal extra gain. Thus, urge researchers consider using whenever one treatments same experiment.