Testing the Shared Knowledge and Skills Hypothesis for Reading-Writing Relation

• In Event: Advancing Understanding of Oral Language, Motivation, Writing, and Shared Skills Between Reading and Writing

Abstract

Objectives
A recent meta-analysis (Kim et al., 2024) has demonstrated a relation that reading and writing. In the present study, we examined the sources of this relation with a particular attention to reading comprehension and written composition, using longitudinal data from English-speaking children who were followed from Grade 2 to Grade 4.

Theoretical framework
Two theoretical frameworks guided this work. The Shared Knowledge Hypothesis (Fitzgerald, 1990; Shanahan & Fitzgerald, 2000) posits that reading and writing are interrelated because they rely on common knowledge, which includes metaknowledge, domain knowledge about content, understanding of universal text attributes, procedural knowledge, and a disposition towards reading and writing. Another theoretical framework is the Interactive Dynamic Literacy Model (Kim, 2020, 2022), which identifies specific skills shared between reading and writing, such as oral language and higher-order cognitive skills. This model also outlines the nature of reading-writing relations, including interactive and bidirectional relations, as well as dynamic changes based on development, linguistic grain size, and measurement.
Although the details vary, both models hypothesize that reading and writing are linked due to their shared skills and knowledge. However, there is limited empirical evidence that simultaneously investigates these shared skills within a single statistical model. In the present study, we examined whether reading comprehension and written composition are related in elementary-grade children, and if so, whether this relation can be explained by shared skills and knowledge.

Method
The data were collected from a longitudinal study following Grade 2 English-speaking children (N = 165, 52% boys) through Grade 4. The racial and ethnic composition of the sample in Grade 2 was as follows: 53% White, 34% African American, 6% Hispanic, 1% Asian, and 4% mixed-race students. In Grade 4, the composition was 58% White, 32% African American, 5% Hispanic, 2% Asian, and 3% mixed-race students. Approximately 68% of the children in Grade 2 and 67% in Grade 4 were eligible for the free and reduced-price lunch program.
The children were assessed on a range of skills in each grade, including reading comprehension, written composition, oral discourse, lexical literacy skills (word reading and spelling), handwriting fluency, higher-order cognition (inference, perspective-taking, monitoring), oral language (vocabulary, grammatical knowledge), and executive functions (working memory and attentional control).

Results
Results from the confirmatory factor analysis indicated that reading comprehension and written composition were strongly correlated, with coefficients of .84 in Grade 2 and .77 in Grade 4. Structural equation modeling revealed that language and cognitive skills predicted both reading comprehension and written composition in both grades, accounting for the relation between the two. An exception was handwriting fluency, which was uniquely related only to written composition, but not to reading comprehension. These findings support the shared knowledge and skills hypothesis.

Scientific significance
The relation between reading and writing has long been established. However, empirical investigations into the shared variance of these skills as influenced by language and cognitive factors have been extremely limited. This study provides empirical evidence to support these explanations, thereby contributing valuable insights to the understanding of the interconnectedness of reading and writing.

Author

• Young-Suk Grace Kim, University of California - Irvine