The Kentucky Instructional Results Information System (KIRIS) exemplifies several key features of current assessment-based reform of education. It relies substantially on "performance assessment" — that is, forms of testing other than multiple choice. It measures student achievement against performance standards that are set higher than current performance, and the stakes are high for schools: financial rewards for those whose KIRIS scores improve and sanctions (soon) for those whose scores do not. As part of a larger study of education reform in Kentucky, RAND staff surveyed teachers and principals across Kentucky to see how KIRIS is affecting their work, student performance, instruction, assessment, and school management. The response was mixed: Although KIRIS provides useful information and encouraged positive changes in instruction, many found it stressful and bad for morale. Most principals found the program burdensome but most also said that the benefits balanced or outweighed the burden. Although teachers reported that expectations increased for all students, the increase was perceived as greater for high achievers than for low achievers or special education students and as more helpful for high-achieving students. Teachers reported using both broad instructional changes and focused test preparation to prepare students for KIRIS. However, they gave test preparation and familiarity much more credit than improved knowledge or skills for initial score gains in their schools. These findings should be instructive for participants and stakeholders in Kentucky schools, as well as in other states contemplating similar reform.