Classroom Practice · CLSM Framework
A Strategy Can Be Research-Based and Still Be Wrong for the Lesson
When somebody tells me an instructional strategy is research-based, I want to know what the research actually measured.
Student achievement? Retention? Reading comprehension? The ability to apply information independently?
Those are different outcomes, and I don't think we should treat them as interchangeable.
The term research-based gets used constantly in education. We see it in professional development, curriculum materials, school improvement plans, and instructional recommendations. Sometimes there's a substantial body of evidence supporting the practice. Other times, we're given very little information beyond the fact that research supposedly supports it.
I want enough detail to judge whether the practice fits what I am teaching.
I use retrieval practice regularly in my classroom. Research across schools and classrooms supports its role in learning, and I have good reasons for including it in my instruction.
But let's say I'm teaching national supremacy in Civic Literacy.
I can give students five recall questions about the Constitution, federal authority, and the Supremacy Clause. Those questions serve a purpose. They provide opportunities to retrieve information students will need later.
However, if the learning objective requires students to analyze a situation involving conflicting state and federal laws, remembering the definition is only part of the work.
Students also need practice identifying which details matter, connecting those details to the constitutional principle, and explaining their reasoning.
I could use retrieval practice every single day and still leave that part of the lesson underdeveloped.
That is why I want to look at what students are expected to do with the information when evaluating a strategy.
We can identify which strategies appear in a lesson plan or during a classroom observation. We can document how often teachers use them. We can even develop expectations around their implementation.
But identifying that a strategy was used doesn't tell us whether it was used appropriately.
I worry that we sometimes spend more time checking whether a strategy is present than asking why the teacher chose it.
Teachers also need to know which parts of an approach they can adjust and which parts need to stay.
I was reading a 2025 study published in Reading Research Quarterly that examined structured adaptations to an elementary content-literacy intervention. Researchers from NC State and Harvard compared the original intervention with a version that allowed teachers to make guided instructional adjustments.
The students in the adapted condition performed slightly better in science reading comprehension and background knowledge. The reported effect sizes were 0.07 and 0.09.
Those are small effects. The study did not find an improvement in general reading comprehension.
The study also involved third-grade students during pandemic-related school closures, so I wouldn't take the findings and assume they apply directly to a high school social studies classroom.
What interests me is how the adaptations were planned.
The teachers weren't given unlimited freedom to change the program. Their adjustments were structured around its instructional goals.
Teachers had room to make decisions while keeping the program’s instructional goals. When schools adopt a program across classrooms, I want that room for judgment to be explained clearly.
Read the 2025 study in Reading Research Quarterly
Teacher preference alone is not a good reason to abandon a practice with strong evidence behind it.
But there also has to be some understanding of the conditions under which an approach was studied, what outcomes it improved, and which parts are essential to making it work.
Otherwise, how do we distinguish a thoughtful instructional adjustment from one that changes the very thing that made the strategy effective?
I have to ask that question about my own instruction, too.
Much of my work involves the relationship between cognitive demands, academic language, lesson organization, and memory. I've spent time studying the research behind those areas and adjusting how I bring them together in my lessons.
Evidence for individual practices does not establish that combining them into my lesson structure produces better outcomes.
I can explain the reasoning behind the decisions I'm making. I can identify research that supports certain components. What I can't do is claim that the complete approach is effective in every setting simply because its individual parts have an evidence base.
That would require additional evaluation.
I want instructional recommendations to make that distinction clear.
The Education Endowment Foundation's 2024 implementation guidance addresses this issue from a school improvement perspective. It emphasizes examining the problem being addressed, considering the school context, and monitoring implementation rather than assuming that adopting a promising practice is enough.
Read A School's Guide to Implementation
Those are the decisions I want more time to discuss with colleagues.
When a strategy is recommended, I'd like to know what it has been shown to improve. I'd like to know whether the evidence comes from similar instructional conditions. And I'd especially like to know how we're supposed to evaluate whether it's accomplishing its intended purpose once we've implemented it.
I don't expect every teacher to sit down and review a dozen research studies before planning tomorrow's lesson. That's not realistic.
But when we're making instructional decisions that affect entire classrooms, departments, or schools, I do think we need to understand more than the name of a recommended practice.
I'm interested in research because I want to make better decisions about instruction. That also means being willing to question how we're interpreting it, including when the research seems to support something I'm already doing.
And if a strategy isn't addressing the learning problem in front of me, I want to be able to recognize that instead of continuing to use it simply because it's considered research-based.
I'd rather be able to explain why I chose a strategy than point to a checklist showing that I used it.
Originally published on LinkedIn on November 11, 2026.
When somebody tells me an instructional strategy is research-based, I want to know what the research actually measured.
Student achievement? Retention? Reading comprehension? The ability to apply information independently?
Those are different outcomes, and I don't think we should treat them as interchangeable.
The term research-based gets used constantly in education. We see it in professional development, curriculum materials, school improvement plans, and instructional recommendations. Sometimes there's a substantial body of evidence supporting the practice. Other times, we're given very little information beyond the fact that research supposedly supports it.
I want enough detail to judge whether the practice fits what I am teaching.
I use retrieval practice regularly in my classroom. Research across schools and classrooms supports its role in learning, and I have good reasons for including it in my instruction.
But let's say I'm teaching national supremacy in Civic Literacy.
I can give students five recall questions about the Constitution, federal authority, and the Supremacy Clause. Those questions serve a purpose. They provide opportunities to retrieve information students will need later.
However, if the learning objective requires students to analyze a situation involving conflicting state and federal laws, remembering the definition is only part of the work.
Students also need practice identifying which details matter, connecting those details to the constitutional principle, and explaining their reasoning.
I could use retrieval practice every single day and still leave that part of the lesson underdeveloped.
That is why I want to look at what students are expected to do with the information when evaluating a strategy.
We can identify which strategies appear in a lesson plan or during a classroom observation. We can document how often teachers use them. We can even develop expectations around their implementation.
But identifying that a strategy was used doesn't tell us whether it was used appropriately.
I worry that we sometimes spend more time checking whether a strategy is present than asking why the teacher chose it.
Teachers also need to know which parts of an approach they can adjust and which parts need to stay.
I was reading a 2025 study published in Reading Research Quarterly that examined structured adaptations to an elementary content-literacy intervention. Researchers from NC State and Harvard compared the original intervention with a version that allowed teachers to make guided instructional adjustments.
The students in the adapted condition performed slightly better in science reading comprehension and background knowledge. The reported effect sizes were 0.07 and 0.09.
Those are small effects. The study did not find an improvement in general reading comprehension.
The study also involved third-grade students during pandemic-related school closures, so I wouldn't take the findings and assume they apply directly to a high school social studies classroom.
What interests me is how the adaptations were planned.
The teachers weren't given unlimited freedom to change the program. Their adjustments were structured around its instructional goals.
Teachers had room to make decisions while keeping the program’s instructional goals. When schools adopt a program across classrooms, I want that room for judgment to be explained clearly.
Read the 2025 study in Reading Research Quarterly
Teacher preference alone is not a good reason to abandon a practice with strong evidence behind it.
But there also has to be some understanding of the conditions under which an approach was studied, what outcomes it improved, and which parts are essential to making it work.
Otherwise, how do we distinguish a thoughtful instructional adjustment from one that changes the very thing that made the strategy effective?
I have to ask that question about my own instruction, too.
Much of my work involves the relationship between cognitive demands, academic language, lesson organization, and memory. I've spent time studying the research behind those areas and adjusting how I bring them together in my lessons.
Evidence for individual practices does not establish that combining them into my lesson structure produces better outcomes.
I can explain the reasoning behind the decisions I'm making. I can identify research that supports certain components. What I can't do is claim that the complete approach is effective in every setting simply because its individual parts have an evidence base.
That would require additional evaluation.
I want instructional recommendations to make that distinction clear.
The Education Endowment Foundation's 2024 implementation guidance addresses this issue from a school improvement perspective. It emphasizes examining the problem being addressed, considering the school context, and monitoring implementation rather than assuming that adopting a promising practice is enough.
Read A School's Guide to Implementation
Those are the decisions I want more time to discuss with colleagues.
When a strategy is recommended, I'd like to know what it has been shown to improve. I'd like to know whether the evidence comes from similar instructional conditions. And I'd especially like to know how we're supposed to evaluate whether it's accomplishing its intended purpose once we've implemented it.
I don't expect every teacher to sit down and review a dozen research studies before planning tomorrow's lesson. That's not realistic.
But when we're making instructional decisions that affect entire classrooms, departments, or schools, I do think we need to understand more than the name of a recommended practice.
I'm interested in research because I want to make better decisions about instruction. That also means being willing to question how we're interpreting it, including when the research seems to support something I'm already doing.
And if a strategy isn't addressing the learning problem in front of me, I want to be able to recognize that instead of continuing to use it simply because it's considered research-based.
I'd rather be able to explain why I chose a strategy than point to a checklist showing that I used it.
Originally published on LinkedIn on November 11, 2026.