RAN Is Not a Measure of Orthographic Processing. Evidence From the Asymmetric German Orthography

In three large samples (N = 1248) of children learning to read German we investigated the correlations between rapid automatized naming (RAN), phonological awareness (PA), phonological decoding (nonword reading fluency), and orthographic processing (word reading fluency and spelling). In a series of hierarchical regression analyses, RAN explained more variance in word and nonword reading fluency than PA, whereas PA explained more variance in spelling than RAN. This pattern was confirmed when PA response times were assessed instead of response accuracy. Two further regression models challenge the view that the RAN-literacy association is mediated by orthographic processing. First, RAN accounts for unique variance in word reading fluency even when differences in orthographic spelling were introduced before RAN. Second, RAN accounts for hardly any variance in word reading fluency when introduced after nonword reading fluency.