Block randomization

Define blocks (e.g., 4 or 6 subjects) and assign equal numbers to each group within each block. Maintains balance during recruitment, avoiding early-closure imbalances. Often combined with variable block sizes to hinder prediction.