Curriculum-based measurement of mathematics (CBM-M) comprises a set of procedures and instrumentation to assess the level and trend of student achievement in early mathematics. The purpose of this article is to review the recent research and psychometric evidence for CBM-M. Although recent developments in CBM-M include procedures to assess early numeracy and application problems, this review focuses exclusively on computation assessment. The results of this review provide evidence that CBM-M is sufficiently reliable and valid for some applications; however, interpretation must be informed by the context and the scope of assessment domain. Mathematics computation is a subdomain of mathematics curriculum and assessment, and therefore, the validity of CBM-M is limited by its construct representation (i.e., stimulus set and task demands). Nevertheless, the review provides support for ongoing development and use of CBM-M as both a general outcome measure and subskill mastery measure for computation. Implications for research and practice are discussed. © 2008, SAGE Publications. All rights reserved.