Consider “direct instruction,” one component–note that I say only “one”–of teacher-centered instruction. Direct instruction has been and is ubiquitous in teacher lessons. For many, but not all teachers, it is the sole means of teaching. The phrase is commonly used by researchers. So why even mention it?
Because there is no one version of direct instruction (see here, here, and here) yet research studies and meta-analyses of Direct Instruction (note capital letters, please) have repeatedly concluded that students exposed to such a pedagogy outperform students receiving other forms of instruction, especially student-centered (e.g., project-based learning, “discovery learning”). Sounds like just another silly intellectual argument when it comes to the linkage between educational practice and research. Not so.
The research evidence that direct instruction (lower-case “d and “i”) and Direct Instruction (capital letters) have positive effects on student learning–as measured by standardized test scores–has been around for decades yet most educational researchers, policymakers, and practitioners who urge school and classroom decisions to be data-driven and evidenced-based have hardly popped champagne corks welcoming this clear direction for what teachers should incorporate in their classroom lessons (see here).
This is puzzling. I need to disentangle different reasons for the arm’s length distance from research on direct instruction. One thread wrapped into this tangle of yarn is that direct instruction is and has been part of teachers’ repertoires since the mid-19th century founding of the age-graded schools. Historically, teachers have lectured, had students recite sentences from textbooks, conducted demonstrations, and assessed students to see that they have learned the content and skills. It is ubiquitous among teachers then and now.
Most teachers have a knapsack of techniques they use with their students that are hybrids of teacher- and student-centered approaches. Blending high student-participation in whole group activity from a discussion to a quiz game, small-group work, and lectures, say in a U.S. history or Algebra 2 class, is common. Mixed strategies of teaching is the norm among elementary and secondary teachers (see here and here).
Yet lecturing, demonstrating, and oral or written questioning to ascertain how much has been learned has gained a bad reputation,often minimized by many observers of teaching even though it is a standard method in most teachers’ repertoires.
And another thread to the puzzle is the confusion between upper- and lower-case direct instruction.
Lower-case “direct instruction” as noted above is prevalent among most teachers although a certain reluctance to admit it exists among most teachers. “Direct Instruction” (upper-case letters), however, has been associated with the original Follow Through experiment involving young children in the late-1960s where one of the pedagogical approaches had teachers using scripted lessons (called Distar or Direct Instruction System for Teaching Arithmetic and Reading). Sigmund Englemann has continually promoted this form of Direct Instruction through generating new curriculum materials and conducting research on its effects on student learning. Other similar programs using scripted lessons combined with other approaches would be Success for All and Open Court reading program.
And there is a final thread caught up in this tangle of yarn. Most university teacher educators and school practitioners–but not all– emotionally lean toward student-centered teaching (and far more student substantive student talk in lessons) but find it hard to implement on a daily basis given the constraints of the age-graded school, district and state curricular demands, testing,and accountability.
Thus, university teacher educators and practitioners accepting direct instruction (lower- or upper-case) openly as an evidence-based mode of instruction because of its superior performance in raising test scores tend to hold it at arms’ length; it would be a betrayal of their beliefs, and here is the kicker, even though on a weekly basis they mix teacher- and student-centered techniques in their lessons.
In my study of teaching methods since the late 19th century, recent observations in classrooms, and many other studies of teaching, it has become clear to me that most teachers (and teacher educators) blend teacher- and student-centered techniques into their classroom repertoires. The mixes will differ by academic discipline, age of students, beliefs about how students learn, and other factors but hybrid approaches are dominant. Thus, researchers have a devilishly tough time in studying the effects of particular teaching strategies because of the blending of both approaches in so many classrooms.
Now here is a puzzle in need of a solution for university researchers, many of whom work in university departments and schools of education. Even with a decided intellectual tilt toward student-centered instruction, university researchers will face decades-old evidence of teacher talk still exceeding student talk in classroom lessons. Such on-the-ground evidence, of course, can not answer the obvious question: what is the ideal ratio between percentage of teacher- and student-talk in a lesson taught using “direct instruction” or whole class discussion; and a lesson taught using small-group activities, cooperative learning, or project-based learning?
Is it 80-20? 70-30? 60-40? 50-50?
And other questions pop up as well: what kinds of teacher talk occur (e.g., asking questions, lecturing on content, demonstrating experiments, reviewing homework)? What are the different kinds of student talk during a lesson (e.g., answering teacher questions, asking their own questions, giving reports, working in pairs and small groups)? What teachers’ verbal moves encourage further student talk (e.g., asking or open-ended question, avoiding “right” and “wrong” responses to student answers)?
Few, if any, researchers, practitioners, and policymakers know answers to these questions.
13 responses to “Evidence To Support Teacher-Centered Instruction (Part 2)”
Hi Larry, thanks for all of this. Folks I talk to in the UK and Australia who are part of these debates like “explicit instruction” rather than “direct instruction” to avoid confusion. The book which we received to implement the Danielson observation rubric specifically says that it uses a constructivist pedagogy emphasizing engagement, with “student centered” practices scattered throughout.
Something I wish was considered more is the research regarding the novice-expert dichotomy and the “expertise reversal effect”. This stems from cognitive load theory–John Sweller writes: “Novices, not possessing appropriate schemas, are not able to recognize and memorize problem configurations and are forced to use general problem-solving strategies such as means-ends analysis when faced with a problem.” Thus, they benefit from explicit instruction until they develop a level of expertise for that field. Once they have, then those techniques used for novices become a burden, with experts thriving more in less structured learning environments. David Didau in England has a good blog post with charts and links here: https://learningspy.co.uk/psychology/novices-become-experts/
I like what David said in his comment (and it’s what I’ve tried to get at in previous comments on Larry’s blog). It seems important, given that teachers use hybrids, to see whether teachers are proceeding along a novice to expert continuum – starting with I do, to we do, to you do – as is a feature of the teacher effects pattern of direct instruction, or if the blend of methods is unintentional or random. A hybrid starting with high levels of teacher- centered support followed by gradual release is supported by cognitive load theory, whereas a hybrid that proceeds without respect to working memory load and expert novice differences doesn’t seem like it would work. A major problem in education would be a hybrid that is essentially random rather than accounting for these important concepts from educational psychology.
I want to thank Zach and David for their comments. Both have gotten me to think more deeply about the notion of “hybrids” that I have used for a long time. Thinking of how teachers move to combine, blend, and mix teaching approaches–does it occur in novice-to-expert fashion over time or is it an unintentional process or even random as Zach mentions?–is something that researchers could explore by observing and interviewing teachers over time. Since classroom research remains scattered and sparse, I am not too hopeful of that occurring but I do thank Zach and David for pushing me to think further about how hybrids of teaching approaches develop.
I’ve been thinking about this, and I think we might disagree on when students achieve the level of expertise. I’m a high school history teacher and I certainly don’t find that my students have achieved expert status by their senior years here. They come to us with a lack of content knowledge, generally mediocre to poor ability to express themselves well in writing, and no research skills. By the time they graduate, we feel we’ve helped them develop all three of these things, so expertise might be at the university level (assuming no slippage). Most of our instruction is explicit throughout all four years, peppered with formative assessments–there are some group projects and other in-class things like pairing, but it’s mostly teacher talk in my department. The chart found in Didau’s blog post I referenced earlier is worth contemplating. To answer Larry’s point on whether this is organic or planned, I think most teachers mix and match on their own, without much thought given to the larger picture.
It depends on what it is. The experiments in the expertise reversal effect are almost all single session randomized controlled trials using, mostly, math materials. The students can achieve relative expert status even by the end of the session, if there are enough trials to teach the material. Something that is 8 elements to hold in working memory at first can shrink to 3 with enough worked examples and practice. But history research isn’t a single type of math problem; it requires an abundance of knowledge of the domain before the student can see patterns in the material and make connections to others like events Etc like an expert.
Thanks for the comment.
The novice-to-expert line of thinking is worthwhile to explore for older students. I’m stuck simply because I am unfamiliar with that literature. I do agree that most teachers “mix and match on their own.”
Pingback: Evidence To Support Teacher-Centered Instruction - kirschner-ED
I had some training “back in the day” – the late 1990s – on Socratic teaching. As others have mentioned, “explicit instruction” is a preferred term, and “I do, we do, you do” is the model. Essentially, you look at an hour of instructional time as an example. The teacher should spend no more than 15 minutes on the “I do” or didactic/explicit instruction. Teachers do need to teach specific concepts. If a child is interested in learning about a historical figure, that child can watch a film or read a book. However, if a child does not know how to read or print their letters, explicit instruction is needed. Since all children are different, this may happen in a small group or with the whole class.
The following 15 minutes is “we do.” The children practice the skill, perhaps with a partner or table group. Maybe they compare their work. In math, they could figure out why some have different answers and where the error was made. The teacher could bring the group back together for discussion and check for understanding before assigning the “you do” work, which could be individual or not, but would take up the remaining 30 minutes.
I am very much pro-autonomy and project-based learning. I taught upper grades where I had some writing and math to teach explicitly, but most basics were already mastered. I think the ideal would be to teach basics as mini-lessons in the early grades and have some projects where students could have more choice. As they progress through the school, students would need fewer mini-lessons and could be given more autonomy in their projects. By high school, they could do projects based on real-world problems of interest or career paths they wish to explore.
Many thanks, Beth, for describing your experiences. “Socratic Seminars,” as I recall the method were the rage in many Advanced Placement and Honors classes in English and Social Studies years ago. I do not know how prevalent they were in American schools. Appreciate your mentioning it.
My Principal loved to have innovators come in and use us as guinea pigs. Since I was young I thought it was great! This was in San Jose, CA at an Intermediate School for 4th-6th (an experiment in itself!).
Thanks for the comment, Beth.
Thanks for additional comment, Beth.