committees have strongly recommended that NAEP incorporate a broader conceptualization of school achievement to include aspects of learning that are not well specified in the existing NAEP frameworks or well measured by the current survey methods. The National Academy of Education panel recommended that particular attention be given to such aspects of student cognition as problem representation, the use of strategies and self-regulatory skills, and the formulation of explanations and interpretations, contending that consideration of these aspects of student achievement is necessary for NAEP to provide a complete and accurate assessment of achievement in a subject area. The subsequent review of NAEP by the NRC reiterated those recommendations and added that large-scale survey instruments alone cannot reflect the scope of these more comprehensive goals for schooling. The NRC proposed that, in addition to the current assessment blocks, which are limited to 50-minute sessions and paper-and-pencil responses, NAEP should include carefully designed, targeted assessments administered to smaller samples of students that could provide in-depth descriptive information about more complex activities that occur over longer periods of time. For instance, smaller data collections could involve observations of students solving problems in groups or performing extended science projects, as well as analysis of writing portfolios compiled by students over a year of instruction.
Thus NAEP illustrates how relaxing the constraint of having to provide individual student scores opens up possibilities for population sampling and coverage of a much broader domain of cognitive performances. The next example is another illustration of what can be gained by such a sampling approach.
The Maryland State Performance Assessment Program (MSPAP) is designed to evaluate how well schools are teaching the basic and complex skills outlined in state standards called Maryland Learner Outcomes. Maryland is one of the few states in the country that has decided to optimize the use of assessment for program evaluation, forgoing individual student scores.4 A population sampling design is used, as opposed to the census testing design used by most states.
MSPAP consists of criterion-referenced performance tests in reading, mathematics, writing, language usage, science, and social studies for students in grades 3, 5, and 8. The assessment is designed to measure a broad range of competencies. Tasks require students to respond to questions or directions that lead to a solution for a problem, a recommendation or decision, or an explanation or rationale for their responses. Some tasks assess one content
Website: <www.mdk12.org>. [June 29, 2000].