The Metropolitan Achievement Test scores for Boston, Massachusetts, have improved from 1987 to 1988, with scores for 8 grades matching or exceeding the 50th percentile as established in 1986. Reading scores improved over the previous year in seven grades. Yet this "good news" provides little information about real reading skills or achievement patterns of students in the Boston Public Schools. Closer analysis of Boston's reading test scores suggests the following: (1) differences in scores between White students and those of Black, Hispanic and Asian students increase as students progress from one year to the next; (2) wide differences in scores between White, Black, and Hispanic students grew even wider between 1987 and 1988; (3) the longer Black, Asian, and Hispanic students remain in the Boston Public Schools, the higher their chances of scoring in the bottom 40th percentile on Metropolitan reading tests; (4) Boston is seriously misusing test score data for policy purposes, particularly in ranking schools; (5) remediation efforts are little more than test preparation classes--the resulting rise in test scores is achieved at significant cost, both in financial terms and in terms of time robbed from real reading instruction; and (6) current reporting of test scores provides no information about what really happens to individual students over time, suggesting that the data is used primarily for public relations purposes, not for identifying the needs of individual students. Recommendations are offered. Data are presented on seven tables and figures. Notes are included. (BJV)