Michael Kane writes in this article that he is in more or less complete agreement with Professor Koretz's characterization of the problem outlined in the paper published in this issue of "Measurement." Kane agrees that current testing practices are not adequate for test-based accountability (TBA) systems, but he writes that he is far from sanguine about the proposed solution strategies. Kane does not expect that innovations in testing methodology (e.g., in test design, equating) will solve this problem; rather, he thinks that it would do better to step back and think about how to address the basic issue that TBA systems are supposed to address: a lack of confidence in the quality of our schools (Cronbach, 1995).