March 26th ~ In-depth: Orthographic Knowledge Meta-Analysis; Growth Mindset & Reading; Identifying Learning Disabilities; Anxiety & RD: Interventions for Severe Reading Comprehension Difficulties
The Weekly Email That Keeps You Informed of the Latest Reading Research!
Welcome to all the new subscribers! This is Issue No. 44
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.
Updates
Please give me feedback on the Recap!
It has been a while since I asked for feedback- just want to see how we are doing. Last week I posted the survey for the recap subscribers, this week I’m opening it up to everyone. Survey here.
Research Highlights
There were some interesting studies this week. I will most likely cover some of these in greater depth in the coming weeks. If you have a preference on which I should cover first let me know by replying to this email.
Intensive Intervention for Upper Elementary Students With Severe Reading Comprehension Difficulties
“This study examined the effectiveness of the intensive implementation of a multicomponent reading intervention (Voyager Passport) with 306 fourth-grade students with severe reading difficulties…Ultimately, students with severe reading difficulties accelerated their word reading and word-reading efficiency achievement, but the acceleration was not enough to also accelerate their reading fluency or reading comprehension.”
Motivational predictors of reading comprehension in middle school: Role of self-efficacy and growth mindsets
“We examined the extent to which self-efficacy and growth mindsets predict reading comprehension growth in sixth grade and the moderating role of initial word reading….Results from a series of latent basis growth models indicated that self-efficacy predicted the initial level but not growth in reading comprehension. In contrast, a domain-general growth mindset did not predict the initial level but did predict growth, which in turn predicted the end-of-year reading comprehension outcome. Unexpectedly, a reading-specific growth mindset predicted none of the reading comprehension outcomes, and the initial word reading skill did not moderate the relation between growth mindsets and reading comprehension growth. Our findings highlight the critical role that self-efficacy and a domain-general growth mindset play in promoting students’ reading comprehension development.”
Impact of Providing Teachers and Principals with Performance Feedback on Their Practice and Student Achievement: Evidence from a Large-Scale Randomized Experiment
“The study found that providing educators with performance feedback had positive impacts on teachers’ classroom practice, principal leadership, and student achievement in mathematics but not in reading.” Note: I reached out the the corresponding author for a PDF, but I could not find any discussion of why they did not find effects for reading. I sent a follow-up email, but they have not responded yet.
Identifying Reading Disabilities: A Survey of Practitioners (open access preprint!)
“To better understand the range of current practices used to identify RDs [reading disabilities] in school-age children, we created and disseminated a survey nationally, and analyzed data from 965 practitioners. The findings indicate lengthy timelines to identify RDs; substantial variability in the composition of assessment teams, identification criteria, and diagnostic labels; and notable opportunities for enhancing practitioner training experiences.”
A Systematic Review of CBM Professional Development Materials: Are Teachers Receiving Sufficient Instruction in Data-Based Decision-Making?
“Results revealed that only a small proportion of information in the CBM instructional materials was devoted to data-based decision-making (12% for presentations and 14% for manuals/books)… Results suggest a need for increased attention to data-based decision-making in CBM professional development.”
The association between anxiety and academic performance in children with reading disorder: A longitudinal cohort study
“The purpose of this study was to examine the association between anxiety and overall academic performance (AP) in children with reading disorder (RD)….Importantly, increased anxiety was significantly associated with reduced AP over time within an individual (adjusted β = −0.22, p = .002). This lends support to screening for anxiety disorders in children with RD. Future research should examine the directionality of this relationship, potential mediators in the pathway and whether interventions to reduce anxiety increase AP.”
First‐Year Teachers’ Informational Reading Instruction: Prevalence, Quality, and Characteristics
“Analysis of lesson segments reveals a range of practices in text‐based instruction, strategy instruction, and representations of content, with the highest quality instruction showing a combination of all three.”
Is orthographic knowledge a strength or a weakness in individuals with dyslexia? Evidence from a meta-analysis
(paper here)
Background
What is orthographic knowledge?
“Orthographic knowledge refers to the information that is stored in memory that tells us how to represent spoken language in written form (from here).”
There is a lot of research documenting phonological deficits in people with dyslexia, but less so about orthographic deficits.
That said, there is strong theoretical reasons to expect deficits in orthographic knowledge if one looks at different models of reading. I am not going to go in-depth here about models or reading (topic for another in-depth section!), but The Dual Route Model, the Connectionist Model, the Simple View Model, Ehri’s phase model, the Reading Systems Framework model, etc. all feature orthographic knowledge as a key component.
There is also some empirical research, but the studies were mixed. There were several studies that showed children with dyslexia showed deficits in orthographic knowledge, but some studies also showed that children with dyslexia had orthographic skills as good as, or better than, the controls. This led to some scientists concluding that orthographic knowledge might even be a strength in people with dyslexia.
Rationale
Given that the prior research was mixed, a synthesis (meta-analysis) could shed more light on the topic than any one individual study. And, there were no meta-analyses that examined orthographic knowledge of children with dyslexia (there was one meta done in adults which found an average effect size d = 1.23).
Purpose of Study
The purpose was to examine if children with dyslexia have an orthographic knowledge deficit when compared with their chronological age-matched peers and their reading level-matched peers (so they ran two models which are discussed separately in the results).
Why use a reading-level match group in addition to same-age peers without dyslexia? Well, “…for a processing skill to be considered a cause of reading difficulties it is not sufficient to show that a group of children with dyslexia perform worse than their general ability and age matched peers, because the difference may simply reflect differences in reading experience (pages 7-11 of this open access paper).”
Methods
Search
They searched several databases and ended up with 68 studies total. These 68 studies contained 80 unique samples for the chronological age group and 33 for the reading-age matched group.
Moderators
Given the mixed findings in the previous empirical research, it is important to examine what factors/variables might influence the effects. So, the authors examined the following moderators:
Level of Orthographic Knowledge (lexical vs. sub-lexical: lexical orthographic knowledge was measured by the Orthographic choice task “Which is correctly spelled: rain-rane?” and sub-lexical orthographic knowledge was assessed with the Word Likeness task “Which looks more like a real word? filk or filv"?)
Type of Orthographic Knowledge score (accuracy vs. response time)
Age (self-explanatory)
Writing System (alphabetic vs. non-alphabetic)
Orthographic Consistency (high: Finnish, Greek, Italian, Spanish, Norwegian, & German; medium: Dutch, Portuguese, Swedish; or low: English, French, Danish, & Hebrew)
Type of Dyslexia (phonological dyslexia, surface dyslexia, or unspecified) If you want to know more about the clinical utility of dyslexia subtypes you should read this paper.
Sample Selection Criteria (they coded this based on how the person’s dyslexia came to be known: standardized tests, teacher nomination, etc.)
Analysis
They used the ‘metafor’ package in r to run the random-effects meta-analysis on a weighted average of cohen’s d effect sizes. Linear models were use to examine the potential moderators.
Results
Chronological age model
The overall mean effect was 1.1742, p< 0.0001, 95% confidence interval [1.0187, 1.3297]. However, after correcting for publication bias, the effect size was 1.1448 (a little lower) but still significant.
What does this mean? Children with dyslexia had less orthographic knowledge than children their same age without reading difficulties
Out of the 7 moderators listed above, only one moderator was statistically significant: level of orthographic knowledge (the effect size for lexical orthographic knowledge was larger than sub-lexical knowledge).
Reading level matched model
The overall mean effect was 0.1811, p< 0.0980, 95% confidence interval [-.0111, 0.3733]. However, after correcting for publication bias, the effect size was .1430 and not significant.
What does this mean? Children with dyslexia had less orthographic knowledge than the sample of children who were matched on reading level
Moderators: None of the 7 moderators listed above were statistically significant in explaining the variability in effect sizes for the reading level-Dyslexia model
Implications
So, what implications does this have for teaching and assessment?
The authors do not make specific proclamations as to what teachers should or should not be doing in the classroom. If you notice in the quote below (from the discussion) they only talk about researchers (emphasis mine):
“Our findings have some important implications for assessment and intervention. Given that children and adults with dyslexia have a significant deficit in orthographic knowledge, researchers may consider including measures of orthographic knowledge when screening children for dyslexia. At the same time, researchers should explore ways to incorporate activities in orthographic knowledge in their intervention programs.”
I wonder how intentional use of the word ‘researcher’ vs ‘practitioner’ is…do they mean that this should be further explored only in researcher-led interventions before being deployed at scale?
Take-home message for practitioners:
In sum, this meta suggests orthographic knowledge is important to assess and teach. But, I think this will not come as a huge surprise (in fact, it might be surprising that the research findings were quite mixed until this meta came out).
Other Notes
(some additional thoughts on the paper)
Causality
What does this meta-analysis say about causality? Does lack of orthographic knowledge cause dyslexia or vice versa?
This study cannot speak to causality and therefore, cannot answer that question, because the study design prohibits such conclusions. Specifically, this study was not an experimental or quasi-experimental study, so it cannot draw strong conclusions as to whether poor orthographic knowledge is a cause of dyslexia, or if poor orthographic knowledge is a by-product of dyslexia.
That said, the comparison of children with dyslexia to both a chronological age group and a reading level match group does try and untangle causal mechanisms. How does it do this? Well “…RL-match designs are commonly used to test assumptions of causality following the logic that if the poor readers perform poorer than RL-matched controls on task A assessing construct B, then construct B is a potential cause for dyslexia. However, if the poor readers only differ from their better reading chronological-age matched peers, the results are considered inconclusive (page 9 of this open access paper on reading level match designs).”
In this meta, the reading level match group had a statistically significant effect size before being corrected for publication bias (a common statistical adjustment in metas), but not after. So, given the discrepancy, I would say that based on this paper, the issue of causality remains an open question.
More on Orthographic Knowledge
The results support the multiple deficit models of dyslexia: or, in other words; that people with dyslexia have multiple deficits, not just phonological deficits. There is converging evidence on the importance of orthographic knowledge in multiple deficit models from computational studies as well (for more information, check out this relatively recent open access paper).
Are the Conclusions Valid?
This is just my opinion, but yes, I think this was a well-designed meta that is tightly linked to the prior literature, is appropriately couched in both theory, and prior empirical research, and included relevant moderator variables. I do not think that the authors overreach in the conclusions they draw.
That said, I do wonder about the overlap/distinction between phonological knowledge and orthographic knowledge in the tasks used. For example, is the orthographic choice task really only measuring orthographic skill? Or, is it somehow also tapping phonological aspects (or other skills). I am not sure, but I did find this recent open access paper, which suggests that my hunch might not be unwarranted. It does not totally undermine the results, but rather, is something to think about going forward.
Stay up to date!
Next week I will be covering some of the studies in the highlights section below in greater depth. These articles often cost $39.95. That’s right, nearly $40 for a single study!
The Recap is only $60 a year (or $5 paid monthly if you just want to try it out at first). So, for the price of 1.5 articles you can get 50+ studies covered in-depth plus hundreds more links to the latest research (many which are open access) delivered right to your inbox! (Note that I can do bulk discounts for groups).
Here’s what people have said about the Recap:
"This looks like it could be one of the BEST possible resources for SoR as it is not just regurgitating the basics and moving things forward. Thank you."
"I have just discovered this blog and think it is fantastic. Allows me access to research I couldn't afford to subscribe to. We don't get enough scientific data at school to base our teaching directions on. Thank you so much"
"I love the recap. It makes it possible for me to keep up with recent research that would otherwise be inaccessible for me. Thank you!!"
"As a time poor teacher (and mum) who is really keen to keep up with the current research your newsletter is a God-send - THANK YOU :)"
"New subscriber, but so far really enjoying all the info!!!"
"I am so appreciative of your work. You are providing a critical service and I know that it is no without much effort. I applaud your efforts and look forward to your continued publications."
"This has been wonderful! Thank you for taking the time to share with us!”