This study describes the development and initial validation of a mathematics-specific spatial vocabulary measure for upper elementary school students. Reviews of spatial vocabulary items, mathematics textbooks, and Mathematics Common Core State Standards identified 720 mathematical terms, 148 of which had spatial content (e.g., edge). In total, 29 of these items were appropriate for elementary students, and a pilot study (59 fourth graders) indicated that nine of them were too difficult (< 50% correct) or too easy (> 95% correct). The remaining 20 items were retained as a spatial vocabulary measure and administered to 181 (75 girls, mean age = 119.73 months, SD =4.01) fourth graders, along with measures of geometry, arithmetic, spatial abilities, verbal memory span, and mathematics attitudes and anxiety. A Rasch model indicated that all 20 items assessed an underlying spatial vocabulary latent construct. The convergent and discriminant validity of the vocabulary measure was supported by stronger correlations with theoretically related (i.e., geometry) than with more distantly related (i.e., arithmetic) mathematics content and stronger relations with spatial abilities than with verbal memory span or mathematics attitudes and anxiety. Simultaneous regression analyses and structural equation models, including all measures, confirmed this pattern, whereby spatial vocabulary was predicted by geometry knowledge and spatial abilities but not by verbal memory span, mathematics attitudes and anxiety. Thus, the measure developed in this study helps in assessing upper elementary students' mathematics-specific spatial vocabulary.

Visuomotor integration (VMI) is the ability to coordinate visual perception and motor functioning. Measures of VMI are commonly used to assess children's readiness for academic learning. Attention and investments towards VMI development are mainly focused on early learners, but some empirical research indicates sustained relations between VMI and academic achievement through middle and high school. To determine the relations between VMI and academic achievement, as well as moderating factors, we conducted a multilevel meta-analysis using a total of 96 articles and 266 effect sizes published over the past 60 years. The pooled effect size revealed moderate correlations between VMI and mathematical (r = 0.39) and reading (r = 0.34) achievement. Educational stage, disability, and intelligence were significant moderators of the relation between VMI and mathematics achievement, whereas educational stage and subdomains of reading skills were significant moderators of the relation between VMI and reading achievement. Implications and future research directions are discussed.