In reading, several curriculum-based measures (CBM) are used for universal screening in Grades K-2. This study summarized the criterion-related validity of early reading CBM tools used in the context of universal screening. After the application of the inclusion and exclusion criteria, 54 studies (53 documents across 34 articles, 16 dissertations, and 3 technical reports) were identified through a systematic review of the literature through December 2018. Forty-eight studies (47 documents across 31 articles, 15 dissertations, and 2 technical reports) were included in the meta-analysis, representing 47,168 students in the US who were enrolled in Grades K, 1, or 2 when they completed the screening measure. A random effects meta-analysis with robust variance estimation was conducted separately for concurrent and predictive correlations to estimate the average effect size between each early reading CBM and the reading outcome measures. Aggregated concurrent Pearson correlation coefficients (r) ranged from 0.343 (phoneme segmenting CBM predicting more complex reading skills) to 0.750 (nonsense words CBM predicting oral reading skills). Aggregated correlations were similar in magnitude for predictive validity. For some measures, correlations were moderated by administration lag. Findings have implications for research and practice regarding the use of universal reading screeners in the early elementary grades.

在阅读方面，几种基于课程的措施 (CBM) 用于 K-2 年级的普遍筛选。本研究总结了在普遍筛查背景下使用的早期阅读 CBM 工具的标准相关有效性。应用纳入和排除标准后，通过对截至 2018 年 12 月的文献进行系统评价，确定了 54 项研究（34 篇文章中的 53 篇文献、16 篇论文和 3 篇技术报告）。 48 项研究（31 篇文章中的 47 篇文献） 、15 篇论文和 2 份技术报告）被纳入荟萃分析，代表美国 47,168 名学生在完成筛选措施时就读于 K、1 或 2 年级。针对并发和预测相关性分别进行了具有稳健方差估计的随机效应荟萃分析，以估计每个早期阅读 CBM 和阅读结果测量之间的平均效应大小。汇总的并发 Pearson 相关系数 (r ) 的范围从 0.343（音素分割 CBM 预测更复杂的阅读技巧）到 0.750（无意义词 CBM 预测口语阅读技巧）。聚合相关性在预测有效性方面的幅度相似。对于某些措施，相关性因管理滞后而有所缓和。调查结果对有关在小学早期使用通用阅读筛选器的研究和实践有影响。