This guidebook shows how school district personnel from many areas--curriculum, instruction, supervision, staff development--can coordinate their activities to focus on instructional improvement. It reveals how data derived from tests, when properly used, can be useful in helping district personnel work with school and community people to assess the adequacy of an instructional program and to improve it. After an introductory overview of the guide, chapter 2 presents a scenario of a week in a relatively small school district where declining test scores have just been released to the newspapers. The ongoing activities, daily crises, and events of the school day form a backdrop against which board members and administrators attempt to assess the significance of the scores and to plan their actions accordingly. Chapter 3 addresses assumptions underlying questions or demands for action from the public in response to release of test scores. These assumptions wrongly assume that a single factor is responsible for low achievement, but actually no single procedure, material, facility, or person operates in isolation from the complex interactions in an educational system. In chapter 4, six districts tell how they responded in a unique way to the problem of increasing student achievement. Each used test scores and evaluative data not only to monitor but to plan instruction. Chapter 5 identifies five reasons why a long-term coordinated district strategy is difficult to achieve. An open systems perspective then shows how changed assumptions can direct a school district toward a coherent plan to interpret testing, evaluation, and instruction. Chapter 6 provides (1) reasons for developing a unique evaluation system, (2) starting points (opportunities and constraints), (3) a sequence for development, (4) selection of emphasis, and (5) methods for identifying sources of support. (TE)