Considering Decodable Texts: Examining Current Evidence and Exploring an Alternative Research Perspective

• In Event: Considering Literacy Learning: Thinking About Texts, Fluency, Vocabulary, and Comprehension

Abstract

One common approach to address the reading gap evident in the National Assessment of Educational Progress (NAEP, 2022) consists of mandates on texts. Legislation in 32 American states now requires primary-grade materials to align with the science of reading (Schwartz, 2023), which often translates into the use of decodable texts. This presentation begins by reviewing existing evidence for decodable texts and then presents an alternative perspective for identifying efficacious beginning texts.
Reviews of Research on the Efficacy of Decodable Texts
Cheatham and Allor (2012) found only three studies where text genre was varied while other elements of instruction, such as curriculum, remained constant. Although initial advantages for decodable groups were observed, this advantage diminished by the end of first grade.
The lack of direct comparisons of text types led Pugh et al. (2023) to analyze text used in early reading interventions. Their analysis of 97 interventions revealed no significant differences between interventions with decodable or nondecodable texts and those with no texts. However, interventions with both text types showed positive effects on word recognition outcomes.
In an update to the Cheatham and Allor (2012) review, seven additional studies were identified (Author, 2023). Results were mixed, with some studies showing associations between decodable texts and improved performance, while others found no significant differences.
An Alternative Perspective on Decodable Texts
As Cheatham and Allor (2012) observed in their review, any text written in English has a degree of decodability. Studies that consider variations in ratios of word types, the unit of decodability, and the volume of text have the potential to clarify the role of different text features in reading acquisition.
First, the quasi-regular orthography of English means that beginning readers are faced with variable letter-sound correspondences in words, especially for vowels (Fry, 2004). Ratios of decodable words to high-frequency words at different points in students’ reading acquisition merit attention. Ehri et al. (2007) found that first graders who read from leveled texts with numerous high-frequency words and words that aligned to a phonics curriculum performed significantly better on assessments than students who received the same curriculum but read from decodable texts.
Second, the current model of decodability in texts, Lesson-to-Text Match (LTTM; Stein et al., 1999), emphasizes individual grapheme-phoneme correspondences (GPCs) as the unit of text design. Once GPCs have been taught in a lesson, words with those GPCs can appear in texts. Evidence suggests that beginning readers also benefit from consistency in orthographic rime units (Treiman et al., 1995).
Third, the amount of text required to learn to read proficiently requires attention. Seidenberg (2017) states that students need exposure to large amounts of different types of texts to generalize orthographic knowledge. Studies are needed of how varying amounts of texts with different features influence reading acquisition.
Scientific research has uncovered numerous aspects of the linguistic system and features of texts that influence reading acquisition. This scientific knowledge merits application in studies that consider how particular features of texts support the acquisition of specific proficiencies in readers.

Author

• Elfrieda H. Hiebert, TextProject