The authors investigated the relatively short-term and longer term effects of grade retention in 1st grade on the growth of mathematics and reading achievement over 4 years. The authors initially identified a large multiethnic sample (n = 784) of children who were below the median in literacy at school entrance. From this sample, the authors closely matched 1 retained with 1 promoted child (n = 97 pairs) on the basis of propensity scores constructed from 72 background variables and compared growth of retained and promoted children using Rasch-modeled W scores and grade standard scores, which facilitate age-based and grade-based comparisons, respectively. When using W scores, retained children experienced a slower increase in both mathematics and reading achievement in the short term but a faster increase in reading achievement in the longer term than did the promoted children. When using grade standard scores, retained children experienced a faster increase in the short term but a faster decrease in the longer term in both mathematics and reading achievement than did promoted children. Some of the retention effects were moderated by limited English language proficiency, home-school relationship, and children's externalizing problems.