Random assignment of experimental units to treatment groups is used to control:

1. Lurking variables threaten any claim that the treatment caused the change. 2. You cannot list every possible lurking variable in advance. 3. The remedy is accomplished by the random assignment of experimental units to treatment groups. 4. Randomisation spreads unknown variables evenly across the groups on average. 5. Sample size and units are chosen deliberately rather than randomised. _Source: OpenStax Introductory Statistics 2e (CC BY 4.0), Ch 1 'Sampling and Data', sections 1.1-1.4_