Six potential problems with the use of randomized experiments to evaluate programs in the field are addressed. Problems include treatment dilution, treatment contamination or confounding, inaccurate case flow and power estimates, violations of the random assignment processes, changes in the environmental context, and changes in the treatment…

The validity of true experiments is threatened by a class of self-report biases that affect all respondents at pretest, but which are diminished by treatment. Four of these inaccurate self-evaluation biases are discussed. Means of detection include external criteria, special conditions of measurement, and retrospective pretests. (TJH)