Abstract
Mathematical performance requires quantitative understanding, linguistic comprehension, and working memory. While children with mathematical disorders may share similar traits, they differ in their specific difficulties. Studies have examined such differences, yet the current research is novel as it also examines reading abilities as a means for classifying mathematical difficulties. A total of 186 elementary-school children (grades 3–4) from seven schools in the north of Israel participated in this study, including 110 who were identified as having mathematical difficulties. Domain-specific skills, acquired mathematical skills, and domain-general skills were assessed. Clustering analysis was performed based on multiple domain-general variables (e.g., intelligence, processing speed, reading) and numerical factors (e.g., symbolic and non-symbolic comparisons). Additionally, a clinical diagnosis of the participants’ mathematical and reading difficulties were conducted prior to the clustering analysis. Five clusters were identified, including low mathematical fluency, mathematical gaps, no difficulties, inhibition difficulties, and low intelligence. The low fluency and low inhibition clusters also included children with reading difficulties. Interestingly, numerical factors only played a small role in determining the subtypes of mathematical difficulties, and only domain-general factors were found to differentiate between subtypes. These findings shed light on the heterogeneity of mathematical difficulties, which are mainly associated with non-specific factors, and highlight the connection between reading difficulties and some subtypes of mathematical difficulties. The findings from this study can be used to develop targeted interventions for children with math difficulties.