Abstract: Analyzing the mathematics and reading performance (measured by average value-added) in Texas' charter schools, the authors of this article find that the performance of charter school students typically begins below that of students in regular public schools in the early years of operation. By the fourth or fifth year of operation, however, they find no significant differences between both types of public schools. The authors also find that the parental decision to exit a charter school is significantly related to school quality and that there is little evidence of systematic differences by family income in the relationship between the probability of exit from a charter school and charter performance. They conclude that "the results provide little support for the fear that lower income parents may have more difficulty acquiring the requisite information on school quality, though it is important to recognize that charter school families are not randomly drawn from the entire school population."