Rapid automatized naming and spelling performance in alphabetic languages: a meta-analysis

The primary goal of this meta-analysis was to estimate the overall size of the relation between rapid automatized naming (RAN) and spelling in alphabetic languages and to examine what factors may moderate their relation. In addition, we examined if the size of the RAN-spelling relation is similar to that of the RAN-reading relation. A random-effects model with robust variance estimation was used to analyze 264 RAN-spelling effect sizes and 385 RAN-reading effect sizes from 103 studies. Results indicated a moderate association between RAN and spelling (r=.35\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$r=.35$$\end{document}) and between RAN and reading (r=\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$r=$$\end{document} .44) with a moderately high degree of heterogeneity. In addition, moderator analyses revealed both similarities and differences in the moderators of the RAN-spelling and RAN-reading relation. Whereas RAN letters and digits correlated more strongly with spelling and reading than colors and objects, RAN correlated more strongly with spelling of real words than pseudowords and more strongly with spelling in opaque orthographies than in intermediate or transparent orthographies. Our findings suggest that RAN is related to both reading and spelling partly because it contributes to the formation of print-to-sound or sound-to-print connections.