Five years of reading comprehension data in Florida Reading First schools were analyzed to address questions regarding student improvement, reduction in the achievement gap, efficacy of site visits to schools making no achievement gains, and effects of student mobility on growth in reading comprehension. Participants were 120,000 students (about 30,000 each in grades K-3) in the 318 schools in the first cohort of Florida Reading First from 2003 to 2008. Outcome measures were the reading comprehension scores on the Stanford Achievement Test (SAT-10) and the Florida Comprehensive Assessment Test (FCAT). The percentage of students on grade level (at or above the 40th percentile) increased, and the percentage of students at high risk (below the 20th percentile) decreased over the five years. Racial/ethnic minority, economically disadvantaged, and English language learner groups improved performance as well, but there was no evidence of narrowing the achievement gap. Reduction in risk for students with learning disabilities was noteworthy. Increased support to low-performing schools was associated with improved performance. Finally, there were significant reductions in growth in reading comprehension associated with leaving a Reading First school. (Contains 4 tables, 4 figures, and 1 note.)