Randomization

is a process that ensures that treatments are allocated at random to the units of a designed experiment. First of all you need a initial assignment of the treatment factor levels to the units. (For example this may be generated in a systematic order, or by use of a design key.) The levels of the blocking factors are then permuted at random, thus generating a random allocation of the treatments to the units. This can be done using the RANDOMIZE directive or the ARANDOMIZE procedure, which additionally allows the levels of the treatment factors to be permuted randomly (useful for example in incomplete-block designs).