Dispelling the myths that MSV is not research supported and that balanced literacy has failed by Dr. Sam Bommarito

Dispelling the myths that MSV is not research supported and that balanced literacy has failed

Dr. Sam Bommarito

Over the past few years, there has been a steady criticism of MSV (using the three systems of information- Meaning, Syntax and Visual).  There have also been criticisms of programs using the constellation of educational practices known as balanced literacy. The criticisms have turned vitriolic. The claim is made that such practices can actually hurt children. One zealot actually suggested that I should be charged with educational malpractice based on the practices I use. Yet the parents of children I tutor are more than happy with the results I’m getting and the children are thriving. They are in the process of becoming lifelong readers and writers. What are we to make of all this? Perhaps the naysayers have missed the mark.

MSV has a place in literacy instruction.

Let’s begin with MSV (the three systems of information).  According to the naysayers, it doesn’t describe how readers actually read. They claim that only poor readers use the three cueing systems. Really? I wonder if they realize that when Marie Clay and others did the groundbreaking research on MSV they looked at what all readers do, not just what poor readers do. As a matter of fact, I think many of the readers of this blog use MSV some of the time when they do their own reading. Let me give you an example. Consider the following:.

“I   b-t    y- -u  c-n   r- -d   th-s”

Back when I was teaching the Analysis and Correction of Reading course, I’d use examples like this to demonstrate that consonants carry more information than vowels. That is important to know. It is especially important to know if you are working with very young readers who sometimes lack the auditory discrimination to hear vowel sounds, especially short vowel sounds. Such children exist. Such children can still be taught to read. But not using teaching schemes that allow only letter by letter sounding.

Let’s get back to the main point. If you decoded the above sample  (and most of you did), you didn’t do it by using only the letter sounds. You used other information, including orthographic information and information from MSV. You likely crosschecked the data from the different systems of information in order to make sure that what your read made sense. BTW- you are good readers using MSV to decode the message. So much for the myth that only bad readers use MSV. Naysayers use quantitative studies to try to prove this point. Qualitative information and qualitative studies are left out. They most certainly should not be left out as the qualitative information provided here clearly shows. Talking about why using both qualitative and quantitative information is important is definitely a topic that will be the subject for future blogs.

There is an excellent review of research around the topic of using context to help reading. Here is a link to the Reading Research Quarterly PDF talking about that. I also include a brief excerpt from that PDF. Regular readers know I’ve talked about this PDF many times.


“….Unlike the position advocated by SOR groups who criticize use of the three-cueing system, Scanlon and Anderson emphasize that the thrust of our argument is that the use of context can be a valuable assist for word solving both when a student’s knowledge of the code is still developing and when in consistencies in English orthography result in only an approximate pronunciation of a word. In either situation, successful word solving that includes a careful interroga-tion of the letters in the word supports the orthographic mapping required for skilled word learning. Additionally, the word-solving process itself becomes generative as it helps to familiarize the student with new letter–sound cor-respondences and orthographic patterns that can then be applied in decoding even more new words, thus expanding the power of the self-teaching mechanism (Share, 1995).This work, which includes studies of K–4 students, showed the effectiveness of the Interactive Strategies Approach, indicating “that using both phonics- and  context-based information facilitates the ability to build sight vocabulary, which in turn enables readers to turn their attention to the most important goal of literacy learning: meaning construction.”

Balanced literacy approaches such as reading/writing workshop and guided reading can and do work in many districts.

I’ve written many times about the manner in which the naysayers criticize the use of balanced literacy practices. Their logic seems to be as follows:

  • The number of children not learning to read is far too high (no argument there- even one kid not learning to read is one too many!)
  • Balanced reading is widely used (no argument there lots of places are using balanced reading in one form or another)
  • Therefore, balanced reading is the reason all those kids are not learning to read (WHOA! A great public relations ploy, but very bad analysis, see below)

Here’s the problem with the third point. It is based on what’s happening in ALL DISTRICTS. That would include districts doing balanced reading with fidelity, districts doing balanced reading but doing it poorly, districts using SOR, districts not really using any particular method, etc.  Based on the fact that SOR is included in what is currently happening,  should we then conclude both Balanced Reading and SOR practices don’t work? When asked this question, SOR advocates maintain that we must look at just the districts using SOR practices and see how well SOR performs.  Point taken. We should. BTW that also means we should look at districts using workshop or guided reading or other balanced literacy programs with fidelity. Then draw conclusions.

That is not at all what is being done currently. Too often the naysayers fail to look at how districts doing balanced literacy with fidelity are doing.  I’ve had them say they can’t do such studies because they can’t really define balanced literacy. Hmm. That can be a problem if you want to conduct research around its efficacy.  It would be especially important if you wanted to do a research study based on a sample of districts doing balanced literacy with fidelity. Definitions of balanced literacy are readily available- see the seminal work of Pressley LINK, or the description of balanced literacy given by Kerns in the most recent issue of The Missouri Reader, LINK (pg. 10)  Any claims that balanced literacy has failed should provide studies that use a carefully drawn scientific sample of districts doing balanced literacy with fidelity. Instead, most of the naysayers base their claims on all districts and the myriad of things one finds when looking at what all districts do.

A few critics who do focus their criticisms on one or more of the balanced literacy programs e.g. workshop or guided reading usually do so when data becomes available showing those programs are working well. When this happens the naysayers use the public relation’s tactic I call, discount and discredit. When positive data becomes available they find ways to “prove” the data is inaccurate or irrelevant.  It doesn’t stop there. In addition, they often provide data to “prove” the SOR practices are better than balanced literacy. These claims often rely on testing instruments like the DIBELS. Those instruments focus mainly on decoding, not comprehension. Overall I find that when I call for research studies to back up the various claims of the naysayers the studies they cite fail to achieve the level of  “gold standard” research.

Let’s talk about  “gold standard”  reading research. Here is a partial list of what one should look for in “gold standard” research.

  • The research reported should have been done over a period of years.
  • The research should involve district wide implementation of a program or select programs WITH FIDELITY. The study would include a careful, precise description of the program being studied. It would include verification that the program was fully and properly implemented (done with fidelity!).
  • It would use measures of reading that include both a decoding and comprehension component. Nell Duke has done work describing what is included in state-wide  reading tests. See my blog for details. LINK  The blog entry includes a chart Duke created describing the content of state wide reading tests (Box 1).
  • The research should be published in a recognized, peer reviewed journalof research. Reading Research Quarterly would be an excellent example of such a journal.
  • The research should be replicable. Ideally should have been replicated several times.
  • Both quantitative and qualitative research can rise to the level of gold standard research. Both can and should be included in the body of research supporting various claims.

Every time I ask the naysayers for proof for the efficacy of their programs or for the proof discrediting the programs they claim don’t work, I don’t get anything close to responses using gold-standard research. Instead, I get stock public relations answers. These answers do not use both qualitative and quantitative information. These answers are often made using limited & limiting views of what reading is and about how research in education should be conducted. These answers are based on research that uses testing instruments that don’t fully test reading.

Moving Toward a Centrist Perspective

My readers are reminded that lately,  researchers in prestigious journals such as the Reading Research Quarterly have concluded that “we are not there yet” in terms of having a science of reading  LINK. I’ve written several times saying that promoting one set of literacy practices and banning all others is a decidedly bad idea. It takes away the right of local districts to find the best fit for their students. It limits the power of teachers to adapt programs to the child. It effectively creates a situation where teachers are forced into teaching situations where they are trying to force square pegs into round holes.

Some have likened the stance taken by selected SOR advocates as more a marketing campaign than a real discussion of research LINK. In the end, the winners in such discussions are the for-profit companies pushing their wares and the losers are the teachers and the kids who eventually find out the hard way that one size really doesn’t fit all.  I’m not against using practices associated with SOR. Regular readers know my students read in decodables. They also are provided systematic instruction in synthetic phonics. But they also read in predictable text and high-interest trade books. They also learn about reading with prosody using materials and ideas provided by literacy leaders like Tim Rasinski LINK. That includes using Rasinski’s materials on prefixes/suffixes and roots and his famous word ladders. Overall,  I try to use programs and materials based on how well they fit the particular child I’m working with. I am not a fan of adopting one size fits all programs. Welcome to the world of the centrist. It is a world in which balanced literacy is carried out with fidelity using the latest and best information about balanced literacy practices. LINK

In Conclusion

Of all the many hats I’ve worn in my 50 plus years in education the one I am the proudest of is that of reading teacher.  I find it upsetting that we seem to be willing to listen to everyone on the planet about the issue of how to teach reading, but seem all too ready to ignore the teachers that are working with the kids directly in the field. It’s been over 50 years since the First Grade Studies concluded that teachers make more difference than any particular program, LINK.  Isn’t it time to stop talking about the combative and shortsighted view exemplified by centering the discussions around terms like “the reading wars” and look instead to anchoring those discussions in more cooperative models like Camborne’s reading quilt LINK (p 5)? Isn’t it time to join what P.D. Pearson once called “The Radical Middle” LINK?  Isn’t it time to stop bickering and to start talking? LINK Isn’t it time to let districts empower teachers by helping them learn a variety of methods so teachers can find and use the methods that best fit each child. Isn’t it time that we finally try out the center? It’s my belief that if we do,  the winners will be the kids. Those kids will become the thinkers and leaders that build that better world that we all continue to hope for.

Happy Reading and Writing

Dr. Sam   (aka the reading teacher in the middle still taking flak from all sides)

Copyright 2021 by Dr. Sam Bommarito. Views/interpretations expressed here are solely the view of this author and do not necessarily reflect the views of any other person or organization.

P.S. If you found the blog through Facebook or Twitter, please consider following the blog to make sure you won’t miss it.  Use the “follow” entry on the sidebar of the blog.

8 thoughts on “Dispelling the myths that MSV is not research supported and that balanced literacy has failed by Dr. Sam Bommarito

  1. Amy Smith

    Thank you for sharing your perspective on MSV and Balanced Literacy. I am a recently certified Structured Literacy Specialist. I also use systematic phonics and decodables with my students. When a student is stuck on a word, I direct them to the letter(s) and sounds (They would have already been taught). Encouraging a student to use any other strategy like skipping the word and reading the rest of the sentence or looking at the first and last letters may work, but it is not as reliable as teaching and reinforcing the English orthographic system (which is 86% decodable). I am speaking from experience having worked with many readers from PK-12. I can tell which students have been taught with balanced literacy and which have been taught with programs aligned with SOR. The former guesses at words (about half the time correctly), and the latter uses letter sound correspondence. When students get to about 3rd grade, these guessing strategies really become a problem because the text starts to become more complex. Students are reading more non-fiction and about topics they have little background on. So using context is not reliable. Using syntax is not reliable. Using the visual information is reliable if it has been taught! Students need strong decoding skills as words get longer and have more affixes, and morphological units. Why not equip them with reliable knowledge they can apply to thousands of words?? Again this is just from my anecdotal perspective.
    (P.S. You can read the message even with missing vowels because you already learned to read and the words are orthographically mapped in your brain).

    1. doctorsam7 Post author

      I’ve been teaching for over 50 years and have taught all of the reading courses usually taken by preservice teachers, and I am familiar with a wide variety of teaching methods. I am currently working with elementary students and have had great success matching kids with what they need. Using the systems of information is not guessing. The idea of guessing appeared in a 1967 article by Goodman. It is not part of how BL folks do things if they are doing BL with fidelity. See the extensive research behind using context found the RRQ pdf. While it is true 86% of the English orthographic system is decodable, it has long been known that significant numbers of the high-frequency words (Dolch/Fry) are not decodable. The idea that context isn’t important is not supported by the research cited in the RRQ pdf. The research cited spans a 25 year period. By the way, I also tell my student to look at the visual information first, but I also teach them to cross-check it with other information systems. Simple example- I read this. Is this read pronounced like it rhymes with seed or like is rhymes with bed. Completely dependent on information from syntax and meaning. I definitely equip my students on how to use orthographical information. I rely heavily on Tim Rasinski’s considerable resources, which he in turn bases on a considerable background. By the way, saying the message can be read because orthographic information may be an explanation but not the full explanation. I’ll end my thought with the question, why not equip them from the get-go on how to effectively use ALL the information present and give them a problem-solving set of skills that help them in multiple settings. Try sounding out antidisestablishmentarianism letter by letter. P.S. The process of reading is complex, not simple. Overall I think Nell Duke’s new model- which builds on the rope model, is the most promising development in developing ways to help children read. BTW- it uses ideas from MANY researchers and many forms of research. It’s a path I highly recommend.

      1. BrIan

        “Using the systems of information is not guessing. The idea of guessing appeared in a 1967 article by Goodman. It is not part of how BL folks do things if they are doing BL with fidelity.”


      2. doctorsam7 Post author

        Really! Expecially note: BL with fidelity NOT the social media version of BL. Properly taught MSV is an ANALYSIS TOOL NOT A CURRICULUM. Teachers look for CROSSCHECKING and BALANCE AMONG USING SYSTEMS OF INFORMATION. In more recent times I have been teaching teachers to teach kids to try sounding first. If that doesn’t work for the child, then the child should check to see if other information may help determine what the world is. BTW just recently one of my twitter followers noted that what Clay taught around using systems of information and what is being taught for set for variabilty dovetail completely . Here is what she said “I see common ground! What Marnie Giesberg as a SoR advocate describes as set for variability & how to teach for it, was also described by Clay 40 years ago. If Ss make an inaccurate attempt at a word, Ts prompt with questions, like Does that make sense? What else could it be?” Another had this to say “SoR privileges phonological processing, Clay’s model teaches for simultaneous integrated processing of phonology, semantics and syntax.” Similtanous processing with crosschecking. Find me classrooms where it is taught that way and still doesn’t work.

  2. Pingback: Balanced Literacy – Reading Recovery Australia

  3. Pingback: Seeking Common Ground and Common Practices: Let’s Promote Dialogue Not Discord in the Discussions Around Teaching Reading Dr. Sam Bommarito | doctorsam7

  4. Steve Dykstra

    Can you name any studies, even 1, in which, non-letter cueing was studied as an instructional variable? There are many studies demonstrating that teaching phonics and other letter/sound skills improves reading, studies that included those skills as variables. Please share similar studies of the other cues. Thank you.

    1. doctorsam7 Post author

      Please review the posting made In the RRQ about the use of context. . “Using Context as an Assist in Word Solving: The Contributions of 25 Years of Research on the Interactive Strategies Approach” by Donna M. Scanlon and Kimberly L. Anderson summary on page 7 of this PDF (link at the end of this comment). Using the context to help in problem-solving words seems to be supported by the considerable research cited in this article. At least 25 years of research results were cited. On a different topic, can you name one study where the balanced literacy approach was studied at a district level over a multi-year time frame, VERIFYING BL was done WITH FIDELITY & showing BL failed to produce positive results in reading achievement? Be sure to explain how BL was defined and what the criteria were for demoing it was done with fidelity. Also, verify that the testing instrument(s) included a significant comprehension component. BTW- About your proposed research design, it would not make sense for a reading researcher to study “Meaning” or “Syntax” as isolated instructional variables. That’s not how MSV even works. I don’t know of anyone who seriously disputes the importance of phonics and letter/sound skills in improving reading, so it’s hard for me to imagine why an experimental design study would ever be conducted in which meaning cues or syntax cues are studied as a variable depriving students of instruction in phonics. As promised here is the link to the RRQ PDF- https://www.literacyworldwide.org/docs/default-source/resource-documents/rrq-sor-executive-summary.pdf?sfvrsn=2561bc8e_6&sfvrsn=2561bc8e_6


Let's talk! What do you think?

This site uses Akismet to reduce spam. Learn how your comment data is processed.