| Summary: | Flexibility or adaptivity in public program evaluation can lead to large savings in time and money, with little or no loss in accuracy, if used properly. In this paper, guidelines are suggested for the employment of classical statistics in adaptive evaluation methodology. Through the case setting of a flu clinic, candidate techniques are demonstrated for handling problems in hypothesis testing, estimation, adaptive allocation of information-gathering resources, and before-and-after-type comparisons. In some cases, classical statistics proves quite adaptable to the requirements of the situation, while in others, its introduction is more artificial.
|
|---|