The Impact of Decoding and Word Recognition Skills on Other Reading Skills and Development

• In Event: Crossing the Decoding Threshold: Research, Remedy, and Repair for Older Students With Word Recognition Needs

Abstract

1. Objectives
The goal of this study is to investigate 1) how grades 5-9 students’ decoding and word recognition (WRD) skills predict vocabulary, morphology, sentence processing, and basic reading efficiency, and 2) how WRD predict growth of these foundational reading skills.

2. Theoretical framework
Decoding is the process a reader engages in to recognize words by applying knowledge in letter-sound correspondence. Recently, the science of reading movement again brings public attention to the fact that many students have not mastered WRD by grade 4 (Hanford, 2018). Wang et al. (2019) discovered a “decoding threshold”: students above grade 4 whose WRD scores were below a threshold almost never achieved satisfactory comprehension. Furthermore, these students also suffered from stagnant growth in reading comprehension. These findings suggest early reading instruction should prioritize WRD skills. However, because WRD is only one of many foundational reading skills to be taught, questions arise regarding the relationship between WRD and other foundational reading skills. This is the focus of the current study.

3. Methods
We conducted longitudinal analysis of students’ test scores on WRD, Vocabulary, Morphology, Sentence Processing, and Basic Reading Efficiency subtests over a period of four years. The dependent variables were time (in years), grade level, and students’ decoding threshold status (applying the rule provided by Wang et al., 2019). The independent variable was scores of each subtest across years. We used R, the statistical software, and the lme4 package (Bates et al., 2015) for the analysis.

4. Data source and materials
The data source of this study was provided by a research contract between ETS and AERDF. A total of 17,133 students from an urban school district on the U.S. east coast provided longitudinal performance on a foundational reading skills battery (Sabatini et al., 2019b) over a period of four years. The battery included WRD, Vocabulary, Morphology, Sentence Processing, and Basix Reading Efficiency. All subtests’ reliability was above .80.

5. Results
Inadequate WRD skills (i.e., being below the decoding threshold) predicted lower performance on all foundational reading skills tested, with effect sizes ranging from -.68 to -.80. Inadequate WRD skills also predicted significantly slower growth rate in all these skills. Figure 1 provides an example for Vocabulary. The same pattern was observed in all other subtests.

6. Scientific significance
These findings suggest that WRD might be a bottleneck for the development of other foundational reading skills. Successful decoding provides the developing reader with opportunities to self-teach the spelling-meaning connection of new words (Share, 1995). This process in turn sets the foundation for building quality lexical representation (Perfetti & Hart, 2002), which is essential for the foundational skills tested in this study. Developing readers who are below the decoding threshold have fewer opportunities to engage in this self-teaching process. In time, they will be left farther behind compared to peers (Figure 1). Early identification of students with inadequate WRD skills and providing them with effective intervention to raise them above the decoding threshold will likely facilitate their growth in other subskills of reading.

Authors

• Zuowei Wang, Educational Testing Service

• Tenaha P. O'Reilly, Educational Testing Service

• Rebecca Sutherland, AERDF