Out With the Old, In With the New: De-Implementation Done Well

It’s normal practice on social media to see a teacher asking colleagues for recommendations for a spelling program. A recent exchange on Facebook stuck with me though. 

In this instance, the teacher was looking for something that would allow the class to study spelling conventions and morphology while providing different levels of spelling words for different students. The inevitable recommendations arrived quickly, with programs, books, resources and systems offered from all directions. This is hardly unusual. Ask almost any large group of teachers how to solve an instructional problem and before long you will have a sizable shopping list. 

I have to admit my first reaction was slightly self-interested. I’ve spent decades studying and writing about spelling, so naturally my urge was to add my own work to the recommendations. Once I got over myself, however, I became far more interested in the premise of the question and asked a little more about what the teacher was trying to achieve and why she wanted another set of spelling lists. 

It turned out her school already had a spelling program. She could see her students were stagnating with it and was frustrated by their lack of progress, but she also was reluctant to depart openly from the mandated approach. She wanted to quietly give her students some more useful instruction before they moved into the next-year level. Here was a teacher paying attention to the evidence directly in front of her and recognizing something needed to change, yet responding to her students rather than continuing with the program felt sufficiently transgressive that she preferred not to draw attention to it. Even then, the obvious route out appeared to be finding another product. And here I was about to add to the pile. So, I did something else. I de-implemented my standard approach in favor of something better. 

I suggested she spend some time exploring spelling instruction more broadly before committing herself to another program. There are excellent resources available, but high-quality spelling instruction rests on deep and broad teacher knowledge about how words are built and how we remember them. A good program can organize and support that knowledge, but no single product can magically provide it on the teacher’s behalf. 

The exchange stayed with me because it captured something I have been thinking about for some time: Literacy reform has changed enormously during the past decade or so, and educators have implemented new programs, routines, assessments, and interventions at an extraordinary rate. Much of that change has been welcome, but implementation has too often meant accumulation. The next stage of reform needs to include a willingness to examine what has accumulated, decide what still earns its place, and become as thoughtful about taking things out as we have been about putting them in.

The first wave

Due to extensive government inquiries into the teaching of literacy in the anglophone world, a wave of reform took place. I call it the first wave, or, in fact, the old wave. The first wave brought some necessary corrections. The role of phoneme-grapheme correspondences received renewed attention, explicit, and systematic teaching returned to the center of conversations about beginning reading, and practices that encouraged children to guess at words from pictures, context, or partial information were increasingly challenged. Phonemic awareness, decoding, fluency, and valid, reliable assessment became routine subjects of professional learning. During that time, educators became far more interested in whether the approaches they were using were supported by research. 

That reform also transformed the education marketplace. Educators can now choose from a formidable range of systematic synthetic phonics programs, decodable readers, phonemic awareness programs, spelling programs, writing programs, assessments, intervention packages, morphology resources, apps, Scope and Sequence documents, and professional learning packages. Many of these are excellent, and I myself write books, develop courses and produce teaching resources. So, this is hardly an argument against educational products. A well-designed resource can save teachers time, provide a coherent sequence, offer carefully selected examples, and help translate research into classroom practice. The difficulty begins when acquiring and faithfully implementing products starts to stand in for acquiring the knowledge required to make good instructional decisions. 

When implementation becomes accumulation

A stark reminder of this came recently when I read an assessment report written by our head of ractice at Lifelong Literacy. A 7 year old student had been referred to us because of concerns about his literacy development. As part of our due diligence, we look closely at the instruction students have already received, since that history often helps us understand why outside intervention has been sought. This particular history was striking because the child had not been deprived of the kinds of instruction currently recommended by literacy reformers. 

He had received an hour each day of a high-quality systematic synthetic phonics program, along with two additional targeted literacy intervention sessions of 15 to 20 minutes each day. He regularly read decodable books, received separate structured spelling instruction three to four times each week, and had explicit classroom writing instruction drawing on established approaches. His intervention was described as data-informed, regularly monitored, and modified, according to his progress. 

I know the phonics program involved here and have completed its training. It is a high-quality program with much to recommend it. What interested me though was the accumulation aspect of it all. This 7 year old was receiving an extraordinary quantity of literacy instruction, much of it bearing all the hallmarks we like to look for: explicit, systematic, structured, sequential, assessed, monitored, and targeted.

Yet the student’s independent production told a rather different story. On a Grade 2 oral reading fluency passage, he read 12 words in a minute and got six correct. His accuracy was 50 percent. His reading fluency was at the fifth percentile and his accuracy at the seventh. His single-word spelling included substantial errors, and on another assessment his regular-word reading was at the seventh percentile while irregular-word reading was at the first. Nonword reading, interestingly, was a relative strength at the 25th percentile. His score on a test requiring him to recognize conventional written words among plausible alternatives also was at the first percentile. 

Perhaps the most revealing information came not from our testing but from the school. He was reported to participate confidently during guided spelling instruction while experiencing difficulty independently applying what he had learned. That distinction between participation during instruction and independent production deserves far more attention than it often receives. 

So many products, so little production

We can count instructional minutes, document intervention sessions, monitor fidelity, and describe an approach as explicit and systematic, but eventually we have to examine what students can independently produce as a consequence of all that activity. We need to know what has become established, what can be retrieved, and what can be applied when the teacher, prompt, scaffold, and worked example are no longer there. 

This becomes particularly pressing in intervention. A student receiving two additional literacy sessions every day is already giving up a substantial amount of time. When progress remains slow despite that level of provision, the idea of simply increasing the dose deserves scrutiny. The phrase “targeted intervention” tells us very little unless the target is derived from close observation of what the student knows, what they produce, and what they need next. 

This also is where the first and second stories begin to meet. One teacher could see her class stagnating under a mandated spelling program and went looking for another program. Elsewhere, a child struggling with literacy had accumulated multiple programs and interventions, yet remained unable to independently demonstrate much of what all that instruction was intended to establish. Neither example tells us that programs are inherently bad. Both prompt a more useful question about what happens when the product becomes the organizing center of instruction rather than a tool in the hands of a knowledgeable and responsive teacher. 

You cannot buy your way into knowledge and experience

I have recently been recording a series about spelling instruction. By the time it is complete, the recordings alone will run more than 20 hours, across 10 two-hour episodes. That is before anyone pauses to investigate a word, follows a word family, tries an idea with students, reads further, or simply spends time thinking about what they have learned. English spelling does not need to be made mysterious or impossibly complicated, but neither can the knowledge required to teach it well be compressed into a Scope and Sequence or a set of weekly lists. The deeper I have gone into the writing system for the past three decades, the more structure I have found, and learning to recognize and work with that structure takes time, study, and experience. 

English orthography draws upon phonology, morphology, etymology, and conventions that arose over time. Teachers encounter the consequences of those interacting elements every time a child asks why a word is spelled as it is, produces an unexpected spelling, fails to recognize a familiar written word, or struggles to retain something that has apparently been taught many times. A predetermined sequence can provide useful structure and carefully selected examples, but it cannot anticipate every word, every error, or every question that will arise in a real classroom. 

Teacher knowledge changes what happens at those moments. It allows a teacher to examine an error rather than simply mark it, to recognize when further repetition is unlikely to supply the missing information, and to decide when a student is ready to move on. It also allows teachers to use programs intelligently, taking advantage of the structure and resources they provide without surrendering professional judgment to them. 

The teacher I encountered online already was exercising that judgment. She could see her students were stagnating and wanted to respond to the evidence in front of her. What she lacked was sufficient confidence in her own knowledge to depart openly from the mandated approach. That is a very different problem from simply needing a better spelling list. 

De-implementation done well

De-implementation is sometimes framed as clearing away practices inherited from an earlier era of literacy teaching. There is certainly still work to be done there, but a mature reform movement must also be prepared to scrutinize practices accumulated during the reform itself. Evidence should help us decide what to introduce, but it should also help us decide what to reduce, adapt, or stop. 

This does not require another pendulum swing, nor does it require teachers to discard systematic instruction and improvise their way through the school day. It requires enough knowledge to judge whether an activity continues to earn its instructional minutes, whether a student has already demonstrated what is being practiced, whether an explanation accurately describes the writing system, and whether an error signals the need for further practice or different information. It also requires sustained attention to student production rather than assuming participation in an evidence-aligned activity is itself evidence of learning. 

Intervention brings these questions into particularly sharp focus because instructional time is limited and the students receiving it are, by definition, already experiencing difficulty. Responsive intervention requires more than delivering a smaller-group or more intensive version of classroom instruction. It requires close attention to what the student produces, coupled with enough teacher knowledge to interpret that production and decide what should happen next. Sometimes greater intensity will be appropriate. Sometimes the better response will involve changing, reducing, or removing something that is currently occupying valuable time.

This is why de-implementation cannot simply become another checklist. Removing something intelligently requires understanding what it was designed to accomplish, recognizing whether it is accomplishing that purpose for the learner in front of us, and having sufficient knowledge to decide what should take its place. 

The new wave

The first wave of literacy reform asked educators to examine the quality of the instruction they were providing. The next wave needs to look more closely at what students are producing as a consequence of that instruction, while deepening the knowledge teachers draw upon when they respond. 

That means retaining explicit and systematic teaching while broadening our understanding of the writing system itself. Orthography is the overarching system in which phonology, morphology, etymology, and conventions interact, and literacy instruction needs to reflect that richness. Reading, spelling, handwriting, vocabulary, syntax, and written expression cannot simply become separate programs competing for minutes in an increasingly crowded timetable. Teachers need enough knowledge to see the relationships among them and enough information to think with when a student does something unexpected. 

There is a pleasing irony in returning to the teacher who began this train of thought. She was worried that departing from her program amounted to “going rogue,” yet what she was actually doing was noticing her students were not progressing and trying to respond. Her instinct to teach them more effectively was sound. What she needed was not another stranger enthusiastically naming a product, but the knowledge and confidence to work out what her students needed from her. 

The next wave of literacy reform can retain the gains of the first while becoming much more discerning about what occupies instructional time. We can examine what has accumulated, remove what has outlived its usefulness, and strengthen the knowledge teachers need to respond to the students in front of them. 

During my EDVIEW360 webinar, I take these questions into the classroom and intervention setting and look more closely at what thoughtful de-implementation can look like in practice, including some practical alternatives that make better use of the instructional time teachers already have. 

That seems to me a much more promising route into the next stage of reform than another shopping list.