The topic of this paper is the second of a two-faceted problem involving achievement testing in the United States. The first facet is the lack of correspondence between test content and intended student outcomes in school districts, and the second facet is "test score pollution." Test score pollution describes instances where test scores for a unit of analysis, such as a class or school, are systematically inflated or deflated without corresponding changes in the content domain that a test is supposed to represent. Test score pollution is associated with standardized achievement tests; however, authentic assessments may be even more susceptible to test score pollution. First, the concept of validity is examined, and second, particular attention is focused on the meaning of school achievement. Third, test score pollution is described and research on the problem is evaluated. To conclude, the effects of test score pollution on limited-English-proficient students are discussed. Responses to the paper by Gary Hargett and Maria Pennock Roma are appended. (VWL)