Introduces the articles of this theme issue focusing on the involvement of key stakeholder groups in the validation of large-scale high-stakes assessments. Each makes a unique but complementary contribution to the understanding of the demands of a comprehensive validation effort. (SLD)

Describes educational assessment in Singapore and discusses changes planned to focus assessment on more real-life situations while taking into account research on learning and critical thinking. (SLD)

An example application of portfolio assessment was developed and the model and estimates of reliability derived from the literature were then used to estimate the characteristics of an operational large-scale portfolio assessment program. Costs were estimated to put results in a realistic context. (SLD)

Provides a general framework for examining the consequences of assessment programs, especially statewide programs that intend to improve student learning by holding schools accountable. The framework is intended for use with programs using performance-based tasks but can be used with programs using traditional item formats as well. (SLD)

Reports on an ongoing evaluation of the Vermont portfolio assessment program. Indicates that the positive news about the instructional effects of the assessment program are in contrast with the empirical findings about the quality of the data the program has yielded. (SLD)

If annual performance assessments are to yield results that can be compared from year to year, many technical problems must be addressed. It is essential that tests to be equated measure the same construct. Methods of equating performance assessment scores, ways of equating system assessments, and standard setting are discussed. (SLD)

An analytic framework is presented for examining the properties and objectives of assessments and scoring systems piloted in a number of state and district testing programs. The article explores the application of this framework, which considers the cognitive components of competence described in studies of expertise, to the analysis of science…

The use of performance assessments in accountability programs is discussed. Performance assessments have the potential to measure important objectives not easily measured by multiple-choice tests. Nevertheless, they are not suitable for every purpose, and their use should not be oversold. Many psychometric issues remain to be resolved. (SLD)

Results are presented from a study of portfolio use for large-scale assessment in the Pittsburgh (Pennsylvania) public schools that are more encouraging for the use of this form of assessment than many other studies have been. Requirements for the psychometric integrity of portfolio assessment are discussed. (SLD)

The development of a hypothetical testing program designed to provide optimum information about student achievement is chronicled. Computer adaptive testing and performance testing are postulated as part of the program. What optimum information really means is discussed. Optimum is only optimal when theory and application are balanced. (SLD)

The extent to which performance assessments can have unintended and undesirable consequences is not yet clear, but preliminary evidence suggests several issues of bias and fairness. Among these is the sheer difficulty of scoring complex performance assessments, as Vermont's experiences with portfolio assessment illustrate. (SLD)