The problem is called assignment bias, and is a threat to internal validity
because it allows two alternative explanations for any differences observed
between the treatments. Specifi cally, it is possible that the scores in one
treatment are higher than the scores in another treatment because there are
real differences between the treatments, or it is possible that the differences
in scores are caused by the fact that the participants in one treatment are
faster (or smarter or older or more motivated) than the participants in the
other treatment.