This current paper uses data collected as part of an efficacy trial funded by a grant from the Institute of Education Sciences (IES) National Center for Education Research (NCER) (See Symposium Justification and Paper #1 for a more complete description of the focus of the broader study). Since participation in the study was voluntary, students showing up for summer school likely differed from students who failed the second semester of algebra but who did not show up in many important ways. If the idea behind credit recovery is to get kids back on track (to recover), how likely is that given how far behind they are? This paper examines which students attend summer school, which students recover the credit during summer school, and how classroom contexts impact the likelihood that various types of students recover credits. Specifically, the current paper seeks to address the following questions: (1) What are the characteristics of students who show up for summer credit recovery, compared with (a) students who don't show up but need to recover, and (b) students who succeeded in Algebra I in grade 9?; (2) Which types of students who show up for summer school are most likely to recover their credits and score well on the post-test in summer school? (a) Students that started far behind in math skills benefit less than students who were far behind? (b) Students who only needed one credit more successful than students who needed multiple?; and (3) How does students' probability of passing summer school depend on the interaction of their individual characteristics and the characteristics of the classrooms they are in (size, teacher qualifications, peer composition [prior academic achievement of students in class])? (a) For example, do students who have failed more classes prior to summer school benefit more from smaller class sizes than their relatively more successful peers? Are students with high numbers of prior failures highly likely to not pass summer school regardless of class size? The author focuses on which students showed up to summer school and will continue to compare and contrast students who showed up versus those who did not in terms of future test scores, course-taking and likelihood of dropout. The current paper uses data from slightly different groups of students for different sets of analyses. For the first of analyses data from all students who attended the study schools and who were first-time ninth graders in fall of 2010 or fall of 2011 were examined to compare the background characteristics and previous academic achievement of three groups of students: (1) students who failed Algebra I and attended one of the study summer school classes; (2) students who failed Algebra I but did not attend one of the study summer school classes; and (3) students who passed Algebra I. The second set examined data from all Chicago Public School (CPS) students who were first-time ninth graders in fall of 2010 or fall of 2011 and who failed Algebra I during their ninth grade year to examine the extent to which students successfully recovered the credit during the summer as well as identify the characteristics of students who did so compared with those who did not recover the credit. The third set of analyses examined data from all students who attended the study schools and who were first time ninth graders in fall of 2010 or fall of 2011 and who failed Algebra I during their ninth grade year to examine how students' probability of passing summer school depends on the interaction of their individual characteristics and the characteristics of the classrooms (size, teacher qualifications, peer composition [prior academic achievement of students in class]) they are in. Conclusions will be able to provide a detailed picture of who attends summer school algebra credit recovery classes, who passes them, and how classroom contexts impact the likelihood that various types of students recover credits.