UC Berkeley

Orthographic depth, the degree of spelling-to-sound consistency in each language, has been hypothesized to affect the ease and effectiveness with which children learn to read (Frost, Katz, & Bentin, 1987). This linguistic factor has been found to have such a powerful effect on the beginning reading process that readers of English (the deepest alphabetic orthography) are estimated to take two and a half years longer to master basic decoding and word recognition skills than the majority of readers of other alphabetic orthographies (Seymour, Aro, & Erskine, 2003). Although the link between orthographic depth and reading acquisition has been well established, there is little to no research examining this relationship beyond the third grade. As a result, it is widely assumed that by fourth grade, reading is the same in all languages. However, because another body of research has established the ongoing and mutually supportive relationship between word reading and reading comprehension throughout development (e.g., Kintsch, 1998; Perfetti, 2007), the current study sought to challenge this assumption by investigating the long-term role of orthographic depth in reading comprehension performance. Hierarchical linear modeling (HLM) was employed to examine subsets of data that include representative samples of typically developing fourth-grade and 15-year-old readers of alphabetic languages from the 2006 Progress in International Reading and Literacy Study (PIRLS) and the 2009 Program for International Study Assessment (PISA). Findings provide tentative indication that by fourth grade, the role of orthographic depth is diminished for competent and skilled readers. However, the poorest 10 to 25% of readers appear to be continually challenged throughout development when there is a greater degree of ambiguity in their script. In contrast, even when students are below average in their ability, they may enjoy the long-term benefits of reading in a shallow orthography. The implications of this study are wide-ranging and include: accounting for orthographic differences in models of reading comprehension, providing more resources for the lowest performing readers of deep orthographies throughout schooling, and acknowledging that reading comprehension instruction and assessment may not be comparable across languages.