A study of school district uses of evaluation and test data is used to illustrate the variety of ambiguities that can appear in qualitative data. Rules of evidence to distinguish valid and potentially nonvalid testimony are proposed. Uses of testimony in evaluation are discussed. (BS)

An alternative approach to Chapter 1 accountability is proposed that would reward schools for reducing the number of students who fail to meet minimum standards on appropriate tests. Program improvement services would be increased and made available to all Chapter 1 schools. (SLD)

The 1988 Hawkins Stafford amendments to the Education Consolidation and Improvement Act required that the national evaluation and reporting system for Chapter 1 identify individual school projects in need of improvement. It is argued that the system is unsuited for this purpose, and suggestions for better indicators are provided. (SLD)

A series of analyses based on a longitudinal study of reading achievement in Chicago (Illinois) illustrate that the conclusions drawn about academic growth and the decisions about types of interventions needed may be very sensitive to the metric that is used (i.e., grade equivalents or item response theory metrics). (SLD)

Options are presented for accountability in Chapter 1 schools and identification of low-performing schools. Two options are suggested that could move Chapter 1 toward more local control or toward more national control. In either case, the shortcomings of using normal curve equivalent gains on standardized tests should be recognized. (SLD)