Universal Screening of Early Numeracy Skills in Spanish-Speaking Kindergarteners

Abstract

This study aimed to validate a fluency-based curriculum-based measurement (CBM) tool for identifying kindergarten students at risk for mathematics learning disabilities within a Spanish-speaking context, addressing the need to examine early numeracy screening beyond predominantly English-speaking environments. A sample of 302 kindergarten students was assessed at three intervals (beginning, middle, and end of the year) using fluency-based measures targeting skills such as quantity discrimination, quantity array, missing number, number identification, and counting aloud. The number identification task and a composite score emerged as particularly strong predictors of risk status, demonstrating adequate reliability, concurrent validity, and predictive validity. Receiver operating characteristic (ROC) analysis and hierarchical linear modeling confirmed the accuracy of this screening tool in identifying students at risk and capturing growth across the year. Longitudinal measurement invariance analyses further confirmed that a stable latent factor was consistently represented across kindergarten benchmarks. These findings are discussed in light of the need for context-specific validation within Spanish-speaking educational settings.