What many educators see as the assistance of a program evaluation is often an intrusion into the governance of the program being evaluated. Redistribution of authority and control occur not only because evaluation findings support one constituency more than another, but the very act of evaluating affords protective status to the initiators of the evaluation and at-risk status to those under review. Most evaluators try to be fair, but within the framework of a complex ideological milieu, conflicts of interest are inevitable, and evaluation both intentionally and subversively changes the way programs are governed. An example of such intrusion can be seen in the evaluation of the Cities-in-Schools program conducted by the American Institutes for Research. The program, established to help troubled students, was not set up for hypothesis testing and lacked managerial uniformity, record-keeping structures, and other evaluation requisites. Various advocacy groups vied for ideological control of the study. Evaluation needs led to standardizing program operations, hiring new managers, and in other ways preempting not only the procedures but also the purposes of the entire program. Because program governance is always affected to some degree by evaluation, evaluators should openly acknowledge their influence and pledge to minimize its effect. (IW)