Joint Modeling of the Component Skills of Reading and Writing

Introduction - The Simple View of Reading (SVR) and the Not-so-Simple View of Writing (NSVW) are well established models of reading and writing, respectively. This study combined and expanded upon these theoretical frameworks of component skills of reading and writing using data from a multivariate meta-analysis of the overlapping indicators and predictors of literacy skills. Method - Data for this study came from a previously conducted meta-analysis of the correlations among component skills of reading and writing (Ahmed, 2013). A synthesized correlation matrix of 77 studies was fit to a joint model of reading and writing. The variables included rapid automatized naming (RAN), phonological awareness (PA), decoding/orthography (D/O), reading comprehension (RC), vocabulary/morphology (V/M), listening comprehension (LC), oral expression (OE), working memory (WM), non-verbal reasoning (NVR), spelling (SP), handwriting (HW), writing quality (WQ), and Curriculum-based (or count-based) measures of writing (CBM). The first class of models replicated and extended the Simple View of Reading. The second class of models was based on components of the Not-so-Simple View of Writing, but did not replicate this model, as the meta-analysis did not identify a well-defined construct for executive functioning. Finally, the last class of models examined the contributions of component skills to both reading comprehension and text generation, and tested whether relations among component skills and writing are mediated through reading. Results - The results supported the validity of the Simple View of Reading as originally proposed by Gough & Tunmer, 1986, that reading comprehension equals the sum of two components (word reading and language comprehension). Results for the models of writing indicated that transcription emerged as an important predictor of writing, and working memory did not explain variance in writing. The models of literacy supported the role of text reading in writing, over and above the role of language and decoding. The results have implications for studying the development of reading and writing in the context of literacy rather than in the context of language.