Editorial: Fountas & Pinnell argue for reducing polarization in reading instruction
Is it time for detente among adocates of different approaches to early literacy?
In Education Week, Irene C. Fountas and Gay Su Pinnell propose that “Teachers, More Than Programs, Make for Great Reading Instruction.” Professors Fountas and Pinnell, who are acclaimed authors of the nearly eponymously named “FPL” community (and materials) for literacy instruction, make a pitch for reducing polarization (their word) regarding reading or literacy instruction.
I find aspects of their argument encouraging. They seem to be offering an olive branch in the “reading wars.” But I am wary of where their terms of surrender lead educators. They seem to cede that early decoding is important. But, they seem to be offering a surrender that declares that their perceived opponents win “on this little tiny piece, but we were right all along.”
Could it be that Fountas and Pinnell are mostly reacting to an earlier report in Education Week that found little scientific support for their commercial offering? Schwartz (2019) described a survey of a nationally representative sample of teachers in which 43% of them say they use materials called Leveled Literacy Intervention authored by Fountas and Pinnell. Schwartz refers to research about early literacy that shows equivocal evidence for the effectives and efficacy of those materials. (That research may be a source for a future post. Fountas and Pinnell also advocate for the “3-cuing” approach. But that, too, will need to be addressed in a later post.)
Now, Schwartz’s analysis is more in the popular press mode, so, perhaps they could dismiss it. Here I argue, however, that the Fountas and Pinnell mostly ignore the evidence about reading aquistion and instruction. Very serious scientists (e.g., Castle et al., 2018) provided thorough analyses about the importanace of learning basic aphabetic principles (and more) as the foundation for fluent word recognition and subsequent expert comprehension.
In this post, I’ll examine topics related to the arguments by Fountas and Pinnell with regard to (a) early decoding instruction, (b) respecting teachers’ authority, and (c) contrasts between reading and literacy. Warning: This post is long…approaching 3000 words; reading the entire thing will require a lot of minutes. Please practice patience!
Emphasis on Early Decoding
Much of the disagreement over reading instruction is predicated on the importance of early, intensive, decoding instruction. This argument has history at least as long ago as books by Flesch (1955) and Chall (1967). Among others, Flesch's popular-press analysis (Why Johnny Can't Read and What You Can Do About It) and Chall's more academic examination of the literature (Learning to Read: The Great Debate) ignited a substantial re-examination of reading instruction. Both books pretty squarely laid the reading deficits of children in the 1950s and 60s in the lap of then current “look-say,” ”sight-word,” “whole-word,” or “meaning-emphasis” methods of instruction and clearly advocated for what we now call “phonics” instruction. (There’s a lot more history than this, but I must defer that, too, for subsequent posts.)
Look-say lost some momentum and phonics gained some mo in the 1960s and 1970s. Publishers hustled to publish curricula that featured phonics activities, with some products better than others. "The education world wants phonics, we'll publish phonics books and materials. Let's GO!" As a teacher in the 60s and 70s, I used some of them…stupid me. Lots of what was published as phonics workbooks and activities was shlock, but some very good materials emerged. However, rather than adopt the very strong curricular packages, many in the reading world slipped right from look-say into “whole language.”
What emerged in many classrooms in the USA was a business-as-usual approach that created an excellent control condition for effective phonics instruction. BAU was some combination of whole-language, basal whole-word or sight-word, and incidental phonics instruction. BAU was, essentially, the comparison condition for lots of studies of phonics; many students got what the schools provided (i.e., BAU) and others got some more explicit phonics program as a “treatment.” With a link to the report of the National Reading Panel, Fountas and Pinnell wrote, “Both classroom- and laboratory-based research have proved the importance of phonics instruction, but such research has not identified any particular kind of phonics instruction to be better than others.” This statement is technically accurate.
Using meta-analysis, the National Reading Panel compared three different types of phonics instruction to BAU. Indeed, in the executive summary for Chapter 2: Alphabetics—Part II: Phonics Instruction, the National Reading Panel (National Institute of Child Health and Human Development, 2000) wrote that the Panel's analysis "showed that effect sizes for the three categories of programs were all significantly greater than zero and did not differ statistically from each other" (p. 2-84). What were those three types of phonics programs? Here's how the panel described them:
(1) synthetic phonics programs which emphasized teaching students to convert letters (graphemes) into sounds (phonemes) and then to blend the sounds to form recognizable words; (2) larger-unit phonics programs which emphasized the analysis and blending of larger subparts of words (i.e., onsets, rimes, phonograms, spelling patterns) as well as phonemes; (3) miscellaneous phonics programs that taught phonics systematically but did this in other ways not covered by the synthetic or larger-unit categories or were unclear about the nature of the approach. (p. 2-85)
The effect sizes for all three types significantly differed from zero (i.e., comparison instruction or BAU) but did not significantly differ from each other. Using the data from the NRP's analysis of outcomes, I created the accompanying figure; it shows the effect sizes for the four conditions (BAU vs miscellaneous, analytic, and synthetic phonics). Readers may draw their own conclusions about which type of phonics they hope the schools their children attend.
But, not every educator wanted to teach phonics. Maybe they didn't feel comfortable with an approach that they, themselves, found confusing (“a rough cough plough me through” or “ghoti'“ as a spelling of “fish” will be familiar to teachers who attended an anti-phonics education program). So, forget accurate translation of the squiggles to normal speech; let's go for meaning!
Well, going for meaning requires that one can read the actual words. If you can not turn the squiggles into words in your native language, you might was well be reading Martian. You are not going to know what it means.
To be sure, you may get hints from other sources. There's a picture of a large, reptile near what you're reading...in the text you see a ‘d’ and an ‘r.’ Do you say “dinosaur” or “dragon?” Either way, it slows you down and confuses you. You kind of stutter. Other kids laugh at your reading. You're too young to know what acronym means, but you think, “‘WTF.’ I must be dumb. I came to school to learn how to read. I'm failing.”
Guessing based on context and one or a few letters is often known as using different cues to determine how to say a word…pictures, context, and the like, but not much from the spelling. (As I noted earlier, the “cuing” system is a topic for another post…but see Hempenstall, 2006, and Schwartz, 2020, if you’re in a hurry.)
Fountas and Pinnell, who may not be the most ardent anti-decoding advocates in education (Frank Smith and Kenneth Goodman, who might be called champions of the whole language approach—'“your guess is as good as mine”—probably still have the lead) but Fountas and Pinnell expressly promote cuing and such.
Teachers Deserve Respect
Fountas and Pinnell predicate part of their editorial on honoring teachers' expertise and professional committment. It's wonderful and valuable to honor teachers. I agree. I recommend doing so!
Along with some other groups of workers in western societies (e.g., nurses, and even before the pandemic), US teachers reported they were overworked and underappreciated (Camera, 2019). It's hard to imagine that teachers in many other parts of the world do not have similar opinions. However, I hear that educators in some Asian cultures, for example, receive respect, and I wish that would be the case in the USA, Canada, Great Britain, and other Western states.
The FPL Web site uses a subtitle about “Elevating Teacher Expertise” and additional argument that amounts to this case: “Experienced teachers know their children and, using their reading expertise, can adapt instruction to meet needs in ways that exceed those by explicit instructional approachs. They do not need scripted lessons because they can do better than those ‘programs.’“
I do like the idea of adapting instruction to meet learners' needs! That idea is central to special education. Let's do it!
But in this case, I heard a variation on an argument from the 1960s about the “teacher variable” explaining more of the variance than the method or program variable. Dykstra (19xx), one of the researchers who conducted an extraordinary study in the 1960s, addressed the ideas that teachers mattered more than programs. The fundamental idea is that, if one has 2, 10, 100, or 1000 teachers using the same program (let's just say Fountas and Pinnell), there will be a distribution of results; students of some teachers will have better outcomes than those of other teachers. Imagine the proverbial "bell curve" of outcomes; the students of some teachers do average, others below average, and others above average.
Sure, part of that variation will be the result of the distribution of the SES of kids, days taught during the school year, absences, etc. Let's presume that we can wash those factors away (and, BTW, using powerful statistics we can). So, just think about a huge number of teachers and their kid's outcomes. With a moment's reflection, it becomes clear that the "teacher variable" makes a difference in child outcomes. Got that? Kids of some teachers have better outcomes. The kids with the lowest teachers were way lower than the kids from highest teachers. Assuming that the kids at the average have scores of, say, 100, then the kids who are with the lowest 34% of the teachers have scores of something like 85. And the kids with the above average teachers have scores of 115! Yikes! That teacher difference is 30 points...an effect size of 2.0! That's huge.
Okay, so hold that normal curve in your head. Now let's make a new normal curve and overlay it on that first one. It'll be the same shape (lowest to highest, skewness, etc.), but it's for a damn clearly more effective program than the first curve. In this case the kids at the average have scores of, say, 110. Now the kids who are with the lowest 34% of the teachers have scores of something like 94-95. And the kids with the above average teachers have scores of 125!
The range of differences in the teachers is still the same. The outcome for kids in the different programs, however, is way different. So, yes, better teachers do better than weaker teachers. But, if they use more effective programs, their students have better outcomes, regardless. When using a stronger instructional approach, the children of weaker teachers, average teachers, and superior teachers all have better outcomes than when using a weaker instructional approach.
Reading and Literacy
Just for a moment, let's think about "reading" and "literacy." Although I suspect that my views about it would cause some concern for Professors Fountas and Pinnell, I agree that literacy is the higher order goal.
If you ask me if I want students to be reading authentic literature, I'll stand up and vote "yes" right now. Feed them "Good Night, Moon" (Margaret Wise Brown), Max's Birthday" (Rosemary Wells), and the "Hunger Games" (Suzanne Collins)! Don't forget Lois Lowrey's "The Giver" or Cynthia Voight's "A Solitary Blue," and (duh) the Harry Potter series.
Here, however, is the simple truth: If you don't know how to read, you are shut out of all those and every other literary experiences. If you do not know how to read, you have no choice about where you can read a bit of authentic literature. If you don't know how to read, you won't be learning what your peers are learning about history, dinosaurs, and space. If you do not know how to read, you might as well act up during reading period, thereby escaping the aversive situation.
So, what does “to read” mean? Well, you have to be able to turn those freaking squiggly lines on the page into language. Sure, you already have language, as Fountas and Pinnell would probably assert. But you have to make the connection between those little bits of (usually black) ink on the (usually) white page so that you can accurately say, “I can read.”
Once you get good at that, then you can do all the wonderful things that the Fountas and Pinnell supporters want you to be able to do.
Fountas and Pinnell, seemingly calling for detente in what has sometimes been called “the reading wars” argue that “Over the decades, beliefs about the 'right' way to teach reading have vacillated widely.” I agree with their statement that the tension is harmful to reading educators and, consequently, the children whom they teach.
Their straw dog is that those who want to teach decoding early, expressly, and effectively are uninterested in literacy, literature, children's interests, and on and on. Those educators are promoting unbalanced literacy.
What counts as “balanced literacy?” It’s not quite clear to me from what Fountas and Pinnell are distancing themselves, other than an argument against labeling practices. There is a certain wisdom in such distancing. But, what do they mean by “balanced literacy?”
Because people can and often do label practice in haphazard and loose ways, regardless of the practices—including balanced literacy—being labeled, labels can be sources of miscommunication. Witness these anecdotes: (a) “Cooperative learning? Of course! I have kids work in groups all the time.” (b) “Oh, yes. I use direct instruction almost every day, but I try not to lecture too long.”
With regard to balanced literacy, it is important to define terms. Pressley et al. indicated that one reasonable definition came from Pressley's book, Reading Instruction That Works: The Case for Balanaced Teaching. “A main message in that book is that excellent elementary literacy instruction balances skills instruction (e.g., phonics, comprehension strategies teaching) and holistic literacy opportunities (reading of authentic literature, composing in response to text)” (2002, p. 1).
BTW, note that writing is included in the Pressley et al. version of balanced literature. I'm glad that it is and I think that teaching encoding and written expression are important compliments to teaching decoding and language comprehension, a view with strong evidence (e.g., Shanahan & Lomax, 1986).
BTW-BTW, given the analysis by Willingham and Lovette (2014), I'm fairly sure educators should invest only a little time in teaching reading comprehension strategies. Teaching strategies such as self-monitoring or summarizing should (and can) be accomplished quickly...and all done, thank you! If we want students to learn reading comprehension strategies, they need to do so in the context of reading authentic (!) content: Expose them to diverse content that helps them build vocabulary, background and general knowledge, content knowledge in multiple domains, and etc.
About the best argument that Fountas and Pinnell offer is in their last sentence: Children need to “learn how written language is connected to spoken language.” That may be the big ball of wax (“print is talk on a paper”) or it may be phonological awareness and phonics. I’ll take both.
As an advocate of early, explicit, systematic instruction in decoding, I welcome Fountas and Pinnell as advocates for explicr teaching of decoding. They are influential advocates, and their advocacy for early, effective decoding instruction would help sway some reluctant reading educators to adopt valuable practices. I am not willing to cede that I and other long-standing advocates of early, explicit, systematic instruction (including, for example, Direct Instruction’s Reading Mastery, which probably is the epitome of their “rigidly scripted phonics approaches”), reject the fun of reading authentic texts. We just want kids to have access to that fun.
I recommend that teachers use instructional packages that allow them to triumph in teaching children to decode and then have access to wonderful written resources. Those teachers will almost surely feel effective.
I greatly appreaciate Professor Emily Solari’s rapid review of this post. She is right there at the tippy-top of reading educators whom I can call on for help. Listen up when she talks!
Resources and References
Bailey, B., Arciuli, J., & Stancliffe, R. J. (2017). Effects of ABRACADABRA literacy instruction on children with autism spectrum disorder. Journal of Educational Psychology, 109(2), 257-285. http://dx.doi.org/10.1037/edu0000138
Camera, L. (2019, June 2019). International survey: U.S. Teachers are overworked, feel underappreciated. USNews. https://www.usnews.com/news/education-news/articles/2019-06-19/international-survey-us-teachers-are-overworked-feel-underappreciated
Castles, A., Rastle, K, & Nation, K. E. (2018). Ending the reading wars: Reading acquistion from novice to expert. Psychological Science, 19 (1), 5-51. https://doi.org/10.1177/1529100618772271
Dykstra, R. (1977). What the 27 studies said. The Reading Informer, 5(1), 11-12, 24.
Fountas, I. C., & Pinell, G. S. (2021, September 08). Teachers, more than programs, make for great reading instruction. Education Week. https://www.edweek.org/teaching-learning/opinion-teachers-more-than-programs-make-for-great-reading-instruction/2021/09
Hempenstall, K. (2006). The three-cueing model: Down for the count? Education News. http://www.ednews.org/articles/4084/1/The-three-cueing-model--Down-for-the-count/Page1.html
National Institute of Child Health and Human Development. (2000). Report of the National Reading Panel: Teaching children to read: An evidence based assessment of the scientific research literature on reading and its implications for reading instruction. (NIH Publication No. 00-4769). U. S. Government Printing Office. https://www.nichd.nih.gov/sites/default/files/publications/pubs/nrp/Documents/report.pdf
Pressley, M., Roehrig, A., Bogner, K., Raphael, L. M., & Dolezal, S. (2002). Balanced literacy instruction. Focus on Eexceptional Children, 34(5), 1-14. https://doi.org/10.17161/foec.v34i5.6788
Robinson, L., Lambert, M. C., Towner, J., & Caros, J. (2016). A comparison of direct instruction and balanced literacy: An evaluative comparison for a pacific northwest rural school district. Reading Improvement, 53(4), 147-164.
Savage, R. S., Abrami, P., Hipps, G., & Deault, L. (2009). A randomized controlled trial study of the ABRACADABRA reading intervention program in grade 1. Journal of Educational Psychology, 101(3), 590–604. https://doi.org/10.1037/a0014700
Shanahan, T., & Lomax, R. G. (1986). An analysis and comparison of theoretical models of the reading–writing relationship. Journal of Educational Psychology, 78(2), 116–123. https://doi.org/10.1037/0022-0622.214.171.124
Schwartz, S. (2019 December 03). The most popular reading programs aren't backed by science. Education Week. https://www.edweek.org/teaching-learning/the-most-popular-reading-programs-arent-backed-by-science/2019/12
Schwartz, S. (2020, December 16). Is this the end of 'three cueing'? Education Week. https://www.edweek.org/teaching-learning/is-this-the-end-of-three-cueing/2020/12
Willingham, D. T., & Lovette, G. (2014). Can reading comprehension be taught? Teachers College Record, 116, 1-3. https://www.tcrecord.org/Home.asp ID # 17701