In 1996, the Chicago Public Schools (CPS) began implementing a new school accountability policy designed to improve student performance by providing a combination of consequences and support to low-performing schools. The center point of the accountability system, the Chicago school probation policy, designates schools as being on probation if fewer than 15% (later raised to 20%) of their students score at grade-level norms on the Iowa Test of Basic Skills in reading. When placed on probation, schools face the consequences of decreased autonomy and the threat of more severe sanctions. At the same time, probation schools receive direct assistance from several different sources through the policy's external support system. The purpose of the support is to assist schools in strengthening their internal operations, raising expectations for students, and improving instruction so as to foster increased student achievement. This report is based on a two-year study of the design and implementation of the school probation policy in Chicago's elementary schools. The authors commend CPS emphasis on providing low-performing schools regular, long-term assistance to improve while noting that data suggest that in the lowest performing schools, current assistance efforts are insufficient to overcome deep problems in educator and organizational capacity necessary to fundamentally improve instruction. Recommendations for strengthening the CPS support policy include: (1) Constraining the market choice; (2) Clarifying roles of support providers; (3) Development of opportunities for learning/sharing among partners and probation managers.; (4) Discouraging schools from developing multiple and fragmented partnerships; (5) Stimulating the development of in-depth, content-based professional development; (6) Connecting assistance to standards, not just tests; and (7) Promoting better use of data in schools by fostering the development and implementation of multiple diagnostic tools. Noting that not all of these recommendations require additional resources, the authors encourage the district to revisit the design of the system to address the identified problems. Document includes an appendix: Description of External Partners Participating in Study. (Contains 32 footnotes, 5 figures and 1 table.) [This paper was written with the assistance of David Wakelyn.]