The reform movement embracing charter schools is based largely on the promise that these autonomous schools will out-perform public schools plagued by bureaucratic administration--an expectation reflected in the federal NCLB law. However, the many state-based reports have been mixed, and previous national studies have suffered from serious methodological shortcomings. In a multi-dimensional analysis of a large and comprehensive dataset, we found charter elementary schools performing at a level beneath those of non-charter public schools, even after accounting for differences in student demographics and school location. In view of this and previous studies, the best current estimate of the performance of charter schools is that any academic advantage is negligible, isolated, or even negative relative to achievement in non-charter public schools. Implications regarding the premise of the federal law are considered.