July 3rd ~ How to Teach Blending; How Many Letter-sounds to Teach Each Week; Components of Reading Comprehension Tests; Sight Words & More
The Weekly Newsletter That Keeps You Informed of The Latest Reading Research
Welcome to the Reading Research Recap, a weekly newsletter featuring the latest reading research published in peer-reviewed scientific journals. The goal of the Recap is to share recent scientific findings and foster an appreciation of science as a way to navigate the world. I try to make this one of the most informative emails you get each week.
Welcome! This is Issue No. 8
“Nothing in science has any value to society if it is not communicated…” ~Anne Roe, The Making of a Scientist
✏️Notes
I’ve been reading a few different books on science communication (if you’re interested: this one, this one, and this one). I’ll be trying out new formats for presenting the research in the next few weeks.
If you have 2 seconds I would appreciate it if you completed this 2 question survey to help improve the newsletter: https://forms.gle/q4S926VVyUMkbpEk8. I will share the results next week.
Here’s What You Will Learn Today
How to introduce letter-sounds to children (fast vs slow)
The best way to teach blending (segmented vs. connected)
The skills that popular reading comprehension tests actually assess
Whether to buy this book
📊Research
The first two studies used random assignment so I put them at the top of the “Research” section.
RCT: Exploring Rate and Complexity of Phonics Instruction
This paper had two studies. In the first study they compared reading outcomes of ”low-skilled” kindergarteners and first-graders in two groups. One group was taught 3 letter-sounds (grapheme-phoneme correspondences: GPCs) a week, while the other group was taught only 2 GPCs a week. They found that the faster rate (3 GPCs/wk) was better and “…didn’t overwhelm the novice alphabet learners.”
Study 2 examined type of GPCs used in instruction. One group of children received instruction in only single-letter GPCs while the other group received “mixed” instruction of both single and multi-letter GPCs. They found that mixed instruction (single and multi-letter GPCs) led to better transfer to decoding, word reading and spelling. Perhaps because many 2-letter GPCs (e.g., ee, oo, sh) are visually distinctive, have continuous sounds, and are more easily blended (speaking of blending see the next study!)
Citation: Vadasy, P. F., & Sanders, E. A. (2020). Introducing grapheme-phoneme correspondences (GPCs): exploring rate and complexity in phonics instruction for kindergarteners with limited literacy skills. Reading and Writing, 1-30. Link
RCT: Comparing 2 Methods to Teach Blending: Connected vs Segmented Phonation
Rationale: We know synthetic phonics is evidence-based, but which method of teaching synthetic phonics is best?
Segmented phonation methods (e.g., Wilson’s Fundations) teach children to convert graphemes to phonemes by segmenting words (‘‘mmm - aaa - nnn”)
Connected phonation techniques teach children to convert by holding continuant sounds (“mmmaaannn”) without breaking the speech stream
Sample & Methods: Kindergarteners who knew letter sounds but couldn’t decode; Independent samples t-test and Mann Whitney U test
Findings: connected phonation worked better, (even for words with stop consonants that cannot be easily stretched out); perhaps because,
“Children in the segmented condition spoke the sounds of successive letters separated by pauses. This caused them to add schwa vowels that had to be deleted from the stops to say the blends. In addition, pausing plus schwa deletion made it harder to remember the sounds they had just spoken when attempting to blend them.”
Gonzalez-Frey, S. M., & Ehri, L. C. (2020). Connected Phonation is More Effective than Segmented Phonation for Teaching Beginning Readers to Decode Unfamiliar Words. Scientific Studies of Reading, 1-14. Link
How Many Exposures to a Word Does it Take for Mastery?
Rationale: To examine both child-level and word-level characteristics that predict mastery of learning a word
Sample & Methods: 1st graders at risk for reading disabilities; crossed random-effects models
Findings: The average # of times it took to reach mastery was 5.65; 1st graders with low reading skill required 2x as many exposures; poor readers tended to rely more on semantic features of words (like imageability)
Citation: Steacy, L. M., Fuchs, D., Gilbert, J. K., Kearns, D. M., Elleman, A. M., & Edwards, A. A. (2020). Sight word acquisition in first grade students at risk for reading disabilities: an item-level exploration of the number of exposures required for mastery. Annals of Dyslexia, 1-16. Link
Examining Component Skills of Reading Comprehension Tests
Rationale: Unlike decoding, reading comprehension (RC) is a difficult construct to measure, and different tests of RC might assess different components of RC skill
Methods & Sample: children in grades 6-8; several standardized tests of RC; They used Dominance Analysis to analyze component predictors
Findings: “…word- and text-reading fluency skills were generally the most dominant predictors of performance…”; working memory was the least important skill; teachers should continue to assess oral reading fluency in older children who struggle with RC
Citation: Clemens, N. H., Hsiao, Y. Y., Lee, K., Martinez-Lincoln, A., Moore, C., Toste, J., & Simmons, L. (2020). The Differential Importance of Component Skills on Reading Comprehension Test Performance Among Struggling Adolescent Readers. Journal of Learning Disabilities, Link
Other Research
Covered in Brief
In other research, The Texas Primary Reading Inventory (TPRI) is valid, reliable, brief early screener. An important thing to remember is that Dyslexia cannot be diagnosed until children are provided explicit instruction. So, the process should go: universal screener -> identify children at risk -> provide at-risk children with explicit instruction determined by reading inventory -> continued progress monitoring -> complete diagnostic assessment (at the end of grade 1) for children who were not responsive to explicit instruction.
📚Books
Splurge or Skip?
I have to say skip this one…
Too Long; Didn’t Read (TL;DR) version of the summary: At $135, and with little practical information, I have to say skip this purchase.
New Book: The Cognitive Foundations of Reading and Its Acquisition
Full Citation: Hoover W.A., Tunmer W.E. (2020) The Cognitive Foundations of Reading. In: The Cognitive Foundations of Reading and Its Acquisition. Literacy Studies (Perspectives from Cognitive Neurosciences, Linguistics, Psychology and Education), vol 20. Springer. Link
Full Review:
I was excited to delve into this book, as Hoover and Tunmer are highly cited, well-known researchers. I think the authors made some important points, but the book was too long and too expensive for the amount of valuable information contained in it.
In chapter 1, the authors state that the rationale for the book is to ground reading research in a cognitive framework. They say a cognitive framework is necessary because research has shown that teachers who know more about reading research/science are better teachers (i.e., their students have better reading outcomes). They also state that research studies, curricula, assessments, and standards are all designed by different people with different theoretical frameworks. I think this last point is an especially good point, but the authors solution of unifying everything under a “Cognitive Foundations Framework” is unnecessarily long-winded and not practical for teachers or parents.
Their solution encourages teachers to conduct “mapping” exercises. They spend several chapters of the book “mapping” standards (Common Core), currricula (SRA), and assessments (DIBELS Next) onto their framework. By “mapping” they mean listing the components of their framework and putting an “x” next to the component if it was covered by the curricula, assessment, or standard. Could their “mapping” be useful in determining which curricula and assessments are covering certain cognitive skills? Sure, but the process seems archaic, time-consuming, and I’m not sure they needed a whole book to describe how to do it.
I did think that the authors make a good point about there not being enough assessments designed for early elementary school years that cover higher-order language skills. This was a point that was also covered by authors in The Reading League’s recent issue.
One last thought
The “Cognitive Foundations Framework” felt to me like it is re-labeling behavioral terms and phrases used in other models and slapping on a new label with the term “cognitive” in it. They do review several other models but there is no logic to which models they chose to include in their review and which they left out. Therefore, at times, it felt like they were cherry-picking models of reading that made their point.
For example, they state that other models do not show the hierarchy or inter-relations between the components and/or lack specific details about the components. However, the Reading Systems Framework, for example, does show the specific components and their inter-relations, however it is not one of the models reviewed by Hoover and Tunmer and I’m not sure why…
Hope everyone has a good 4th of July weekend (if you are in the US)!
-Neena
Thanks for putting this together and summarizing these results and information! I am a little concerned by the connected phonation/continuous sound blending recommendation. That is often used as a last resort for students with reading disabilities due to the sacrifice it has on spelling skills when compared with segmenting sounds. I would be interested if the study mentioned that connection or isolated the results to students’ ability to decode words.
Thanks for these findings. I noticed that the connected phonation is better and helps kindergartners to pronoun words correctly and read better. This is because they sound the word without breaking the stream thus coming out with the word clearly. This method is very effective in teaching reading.