Stability of early number sense competencies for predicting mathematics difficulties
Learning and Individual Differences (2025). Click here to go to journal page with author, etc. info.
It is my current working hypothesis that measures of number sense (and other Gq abilities such as magnitude recognition) may measure what developmental cognitive psychologists consider part of a quantitative “core knowledge system”. Click here for access to an excellent article that discusses all core knowledge systems. Perhaps these are fundamental cognitive mechanisms that lie beneath the narrow ability stratum in the CHC model of human cognitive abilities.
Abstract
Significant individual differences in children's number competencies exist in early childhood, and these competencies can have long-term implications for academic and economic success (National Research Council, 2009). The present study assessed the classification accuracy of the Screener for Early Number Sense (SENS), a screening tool grounded in an evidence-informed conceptual model of number sense that is designed for children in pre-kindergarten (pre-K), kindergarten, and first grade. A cohort of 450 children was assessed on the SENS, followed by a standardized mathematics achievement measure one year later. Receiver operating characteristic (ROC) curve analyses indicated that the SENS effectively differentiated children who went on to show math difficulties (MD) from those without MD. Accuracy rate was acceptable for pre-K, kindergarten, and first grade. An analysis of SENS performance over time categorized by MD risk status revealed differing patterns of growth across the pre-K vs. kindergarten year. These collective findings underscore the importance of math screening as early as pre-K.
Educational relevance
Educators need reliable and valid tools to identify young children who may be at risk for mathematics learning difficulties. The present study presents a number sense screener that accurately predicts pre-kindergarten, kindergarten, and first grade students' later mathematical difficulties. The study demonstrates how screener results and prevalence rates can be used to identify individual-level student risk and to make subsequent instructional decisions across varying schools or districts.