Reading fluency and comprehension in a transparent orthography: Evidence from Turkish children

Abstract

This study was designed to explore the concurrent relationships of reading skills (i.e., word-level reading fluency and comprehension) with cognitive and linguistic variables (i.e., rapid-automatized naming [RAN], phonological awareness [PA], phonological memory [PM], morphological awareness [MA], orthographic knowledge [OK]), processing speed [PS], and vocabulary), along with parental education level. The interplay among these variables was also investigated. A total of 92 Turkish-speaking children in grades 2 and 4 were administered a battery of tests to measure their performance with respect to reading skills and linguistic and cognitive variables. A preliminary model of reading that illustrated the proposed direct and indirect (i.e., mediating) relationships among the variables was developed. The model was then tested using a classical path analysis which relied on a series of simultaneous multiple regression. The results revealed that alphanumeric RAN and OK remained the strongest and most persistent predictors of fluent word reading at both grade levels. Other variables, namely PA, PM, MA, OK, PS, vocabulary, and parents’ education level, made no significant direct or unique contribution to word reading fluency. PA and PS indirectly influenced second graders’ fluent word recognition through OK. However, only PS made an indirect contribution to word reading fluency via OK in grade 4. As for the reading comprehension skill, fluent word recognition, vocabulary, and parental education level were identified as independent precursors across grades. In addition, the mediating roles of vocabulary and fluent word recognition were emphasized in reading comprehension.