Problems with conducting randomized field experiments in education are explored. Focus is on problems encountered while evaluating a group of dropout prevention projects. Project planners were asked to manipulate the subject eligibility criteria until they identified as eligible three to four times as many students as they could serve. They were instructed to recruit all of them, allowing for the possibility that between 25% and 50% of the those eligible would not apply. A lottery would be used to identify students to be served. Students not selected for the intervention might display "resentful demoralization," acting as though some negative intervention had occurred. Techniques to alleviate this demoralization are discussed. Other problems of attrition in subject numbers included non-random group reassignment by a project director, and failure to track students who left the school or the area. It is concluded that implementing randomized control group designs is difficult and that a perfect experiment of this type can not be implemented. However, with care, the approach can be managed. (SLD)