Washington, D.C.: National Academy Press. Available: http://www.wcer.wisc.edu/nise/publications/briefs/vol%5F1%5Fno%5F3/index... [July 6, 2000]. 147-201). Evidence from California. Education Week, p. 3. 31-92). This is not the place to analyze in detail what ought to be done to reduce the patterns of nested inequalities and concentrated harms in public schooling; any serious policy change is enormously complicated, particularly in the diffuse and decentralized world of public schooling. I conclude with the broad outlines of what would be necessary to reduce, even if we can never eliminate, class (and racial) disparities in American public schools. Ninety percent of the children in this school are poor, 40 percent have limited English proficiency, many move frequently. The very poorest Americans have become even more concentrated; the 100 largest cities’ share of the nation’s welfare recipients grew from almost 48 percent in 1994 to over 58 percent in 1999 (Allen & Kirby, 2000). The deepest problem, then, is that too many students are poorly taught, and students in low ability groups – disproportionately poor students, who are disproportionately of color -- usually are the most poorly taught of all (Good, 1987; Ingersoll, 1999; Weiss, 1997). America’s top income bracket is more likely than other groups to be Jewish, while the lowest bracket is more likely to be Jehovah’s Witnesses. Additionally, higher status people are more likely to hold political positions than lower class people. It is unlikely that a parent chooses to move to a lower-spending district if she can afford to live in a higher-spending district, and districts never vie to spend less in the endless disputes in state legislatures over funding formulas. Oakes, J., Gamoran, A., & Page, R. (1992). In nations with high levels of fertility, upper class individuals tend to have more children than their lower class peers, while in nations with low levels of fertility, upper class families exhibit even lower fertility than average. The variable effects of high school tracking. National Center for Education Statistics. Education Trust. Wealthy, well-educated Americans are more likely to vote and to donate money to politicians than lower class individuals are. Monk, D., & Rice, J. Washington D.C: Brookings Institution Press. Such educational inequality is further reinforced by legacy admission, the preference given by educational institutions to applicants who are related to alumni of that institution. Prodded by this lawsuit, the College Board set up a program to ensure that all public high schools offer AP courses within a few years (currently 40 percent do not), and some schools are encouraging more students to take them (Viadero, 2001). ), Divided we fail: Coming together through public school choice (pp. Abramson, A., Tobin, M., & VanderGroot, M. (1995). Cambridge MA: Harvard University, Department of Government. (1999). Remarks by the president in submitting education plan to congress. It begins with states. In the worst-performing municipality, 49% of the class of 1995 dropped out during the four years before graduation; in the best performing community the drop-out rate was 0%. Iatarola, P., & Stiefel, L. (2003). Social class is not significantly correlated to religiosity, an index of how strongly religious a person is. These are, however, the fields for which the relationship between subject area knowledge and effectiveness has been most clearly demonstrated (Goldhaber & Brewer, 2000). Students with disabilities or with limited English proficiency are not likely to be in high tracks regardless of their talents (August & Hakuta, 1998; McDonnell, et al., 1997).