Testing of students and computer systems to store, manage, analyze, and report the resulting test data have grown hand-in-hand. Extant research on teacher use of electronically stored data are largely qualitative and focused on the conditions necessary (but not sufficient) for effective teacher data use. Absent from the research is objective information on how much and in what ways teachers use computer-based student test data, even when supposed precursors of usage are in place. This paper addresses this knowledge gap by analyzing the online activities of teachers in one mid-size urban district. Utilizing Web logs collected between 2008 and 2010, I find low teacher interaction with Web-based pages that contain student test information that could potentially inform practice. I also find no evidence that teacher usage of Web-based student data are related to student achievement gains, but there is reason to believe these estimates are downwardly biased.

This paper estimates the causal effects of enrollment in one of five oversubscribed high-quality Los Angeles charter schools using a lottery design. Enrolling in a charter school increased 11th-grade standardized test scores and enrollment and persistence in four-year colleges substantially, but had no effect on high school GPA. Charter students took more advanced coursework, were less likely to skip class, were more likely to apply to a four-year college, and reported more teacher support for college-going. The increase in four-year persistence appears to be driven by more enrollment and persistence in University of California (UC) campuses, which have higher graduation rates than the likely alternatives. This suggests that the effects of charter schools could depend on the availability of high graduation rate colleges.