Numerical estimation strategies are correlated with math ability in school-aged children

Abstract Various laboratory-based numerical cognition measures are often found to be linked with standardized math scores, yet the exact nature of these links over development remains in question. In this study, we investigated how 0–100 number line estimation performance and dot numerosity discrimination explained variance in 6- to 8-year-olds’ standardized math ability scores, and we assessed links between tasks. Our primary goal was to assess the link between estimation performance and math ability from a theoretical perspective that models number line estimation as an instance of proportion estimation. As age increased, children’s proportion-based strategies for number line estimation became more sophisticated and estimation bias decreased. Children who spontaneously inferred and used a central reference point to guide their proportion estimates scored higher on the math ability measure (TEMA-3), demonstrating a relation between more advanced proportion estimation strategies and higher math abilities. Individual differences in estimation bias also made a lesser contribution to explaining variance in children’s math ability. Dot discrimination accuracy uniquely contributed to children’s math ability. Number line estimation performance and dot discrimination accuracy were not associated, consistent with prior research suggesting that number line estimation does not directly map onto mental numerical magnitude representations. Educational implications and future directions are discussed.