When you say "can be internally valid" what do you mean?
http://en.wikipedia.org/wiki/Internal_validity
What about interactions from repeated treatments?
?
Also, how do you know the selection bias does not create non-causal explanations for observed dependence?
I don't understand your hypothetical. Could you give a concrete example?
I don't understand your hypothetical. Could you give a concrete example?
http://www.maths.bris.ac.uk/~maxvd/didelez_etal_StatSci_final.pdf
With repeated measures design the problem is whether the "washout period" is sufficient.
http://www.johndcook.com/blog/2013/02/04/four-hours-of-concentration/
And since this is the Internet, and facts are involved, our gwern turns up there also.