In a repeated-measures experiment, a researcher is concerned that completing problem solving task A may influence the participant's performance when solving problem B. Therefore, he randomly assigns half the participants to complete the A task first, and the other half complete the B task first.
This technique is called