This report outlines the major themes of the 1999 annual conference of the Center for Research on Evaluation, Standards, and Student Testing (CRESST), which focused on creating benchmarks for accountability in education. Opening the conference, a panel of experts focused on recent work regarding standards, quality, and fairness in the assessment field, with a summary of CRESST's work and a discussion of accountability issues presented by Eva L. Baker. The national context for assessment was discussed by Joseph Conaty, director of the National Institute on Student Achievement, Curriculum, and Assessment. As part of this panel, Richard Elmore reviewed the recent report on Title I Testing. Another session focused on testing and accountability for high stakes purposes, featuring presentations by Lorraine McDonnell, Edward Haertel, and Lorrie Shepard among others. The third panel discussion discussed the validity of testing and assessment of all children, and another panel considered the impact of accountability systems on instruction and outcomes. Small group discussions, in concurrent sessions held each day of the conference, focused on: (1) technology use; (2) the implementation of assessment in large urban school systems; (3) reporting school and district progress; (4) using multiple measures of assessment; (5) measuring school performance through the California Academic Performance Index; (6) issues surrounding college entrance; and (7) systemic and classroom perspectives on the alignment of goals, instruction, and the use of assessment. In the closing session, Robert Mislevy discussed making sense of the data from assessment systems, and other presenters continued with the discussion of testing issues and validity concerns. (SLD)