In addition to the threat of assignment bias, a between-subjects design must also be concerned with threats to internal validity from environmental variables that can change systematicaiiy from one treatment to another (Chapter 6 ,p.180) Thus there are two major sources of confounding that exist in a between-subjects design.