Recognizing the potential of accountability tests to alter classroom instruction, an overview is provided of the federal government's past influence on educational assessments. The past 50 years has witnessed the function of federally engendered educational assessments shift from monitoring the use of federal funds for programs prescribed for statute-specified student populations to assuring the academic achievement of all students. Federal influence over state-level accountability testing has expanded recently because of key federal legislation enacted in the early years of the 21st century. The author suggests a framework for rethinking an appropriate federal role in U.S. educational testing based on two dominant questions: (1) What level of control should the federal government have over educational accountability tests? and (2) What should be the measurement mission(s) of those tests? Control options for accountability tests, ranging from zero federal control to total federal control, are presented. Consideration is given to three design dimensions that will govern the degree to which an accountability test is apt to have a beneficial impact on instruction, accountability, or curriculum. It is concluded that, especially through the various reauthorizations of the Elementary and Secondary Education Act of 1965, not only has the federal influence on the nature of U.S. accountability tests intensified during the last half-century, but the range of students affected by these tests has also expanded significantly. Although federal education-related laws have impacted both instructional practices and curriculum, to date the ensuing accountability tests have been designed to support accountability functions rather than instructional or curricular initiatives. The author advocates that an appropriate level of federal control and the proper measurement missions for important accountability tests, should frame any serious rethinking of the federal role in educational assessment. (Contains 2 footnotes and 2 figures.)