Date of Award
Janice M. Keenan, Ph.D.
Listening comprehension, Quantile regression, Reading comprehension, Vocabulary, Word decoding, Working memory
Many studies to date have examined cognitive factors that drive individual differences in reading comprehension. However, these studies often focused on typical readers, and it is not clear whether their findings apply similarly to readers performing in the extreme ends of the distribution, i.e., poor and good readers. In this dissertation, we used quantile regression on a sample of 834 children (age 8-18) to advance our understanding of the relative importance of different component processes of comprehension not just for the typical but also for poor and skilled readers. In Study 1, we examined how the relative importance of components of the Simple View of Reading, namely word recognition and listening comprehension, might vary across different skill levels of reading comprehension. Because there are large differences between tests in the component skills they assess, reading comprehension is defined by five different tests. This is to determine how generalizable our findings are across tests. In Study 2, we deconstructed listening comprehension into vocabulary and working memory to see whether their contributions to reading comprehension beyond decoding skills also vary across reading skills. In Study 3, we determined whether the contributions of vocabulary and working memory found for reading tests generalize to listening tests.
We found that, for three out of five reading tests, the contributions of the component processes vary as a function of reading performance levels. Therefore, the results previously found for typical readers are not always generalizable to poor and skilled readers. Additionally, working memory is a reliable component of listening comprehension only for some reading comprehension tests whereas vocabulary is a much more robust component of listening comprehension across all reading comprehension tests and readers. Finally, we found that reading and listening comprehension rely on the same language processes of vocabulary and working memory once differences in decoding skills are taken into account. Interestingly, we also found some evidence that working memory may be more influential in reading comprehension than in listening comprehension. We discuss the implications these findings have for diagnosis, instruction, and research.
Copyright is held by the author. User is responsible for all copyright compliance.
Hua, Anh Ngoc, "Quantile Regression Analyses of the Component Skills in Various Comprehension Tests" (2016). Electronic Theses and Dissertations. 1180.
Received from ProQuest
Anh Ngoc Hua