Explicit and direct instruction: what they are, how to teach them, and what the research says

What it is

Two phrases get used for this, and it helps to keep them apart.

• Explicit instruction (or direct instruction with small letters) is a way of teaching any lesson: the teacher states the goal, explains and models in small steps, guides students through practice with lots of questions, checks that every student has it, and then gives independent practice, with review in later lessons. It's often summed up as "I do, we do, you do".
• Direct Instruction (with capitals) is a specific family of scripted programs in reading, math and language, with carefully sequenced lessons, choral responses and placement tests. Its research base is the meta-analysis by Jean Stockard and colleagues.

Most of what follows is about explicit instruction, which any teacher can use with any curriculum. Where the research is specifically about the scripted programs, we say so.

Barak Rosenshine's summary for teachers lists ten principles that describe explicit instruction well: begin with a short review; present new material in small steps with practice after each step; ask a large number of questions and check every student's response; give models; guide practice; check for understanding; aim for a high success rate; scaffold difficult tasks; give extensive independent practice; and review weekly and monthly. He drew them from three bodies of research (cognitive science, studies of effective teachers and research on scaffolding) and noted that they agree.

Explicit doesn't mean long. A good explicit lesson has more student talk and practice than teacher talk: short explanations, many questions, and students doing the work as soon as they can.

Why it works

Working memory is small. A student meeting a new idea can hold only a few new pieces in mind at once, and if they're also trying to work out what to do, there's little room left to learn it. Clear explanations, worked examples and practice in small steps reduce that load. Paul Kirschner, John Sweller and Richard Clark set out this argument in 2006: for novices, minimal guidance is less effective and less efficient than strong guidance, and the advantage of guidance shrinks only once learners know enough to guide themselves.

The meta-analysis by Louis Alfieri and colleagues is the most direct test. Across 164 studies, explicit instruction beat unassisted discovery under most conditions. But discovery with support (feedback, worked examples, scaffolding, asking learners to explain) beat other forms of teaching. The lesson isn't "always lecture". It's "don't leave novices to work it out alone".

For students who struggle, the evidence is especially clear. The What Works Clearinghouse's 2021 guide on math intervention in the elementary grades rates "provide systematic instruction during intervention" as Strong: clear models, a planned sequence, practice with feedback and cumulative review. Its other five recommendations, also rated Strong, describe the content of that instruction: precise math language, concrete and visual representations, the number line, deliberate teaching of word problems and timed fluency practice.

The scripted Direct Instruction programs have a large body of research behind them. Stockard and colleagues' meta-analysis covered 328 studies from 1966 to 2016: every estimated effect on academic outcomes was positive and statistically significant, the effects held up after the programs ended, and they were larger with more exposure.

How to start, by grade

Examples

• I do, we do, you do with a new soundKindergarten–1st grade

  I do: "This is sh. It says /sh/. Watch: /sh/ /i/ /p/, ship." We do: the class blends three sh words with you, then without you. You do: each child reads two sh words to a partner while you listen to three or four children. If most hesitate, back to "we do" with two more words.

• Worked example, then your turnGrades 3–8

  Solve 3/4 + 1/8 on the board, saying each step and why. Then put up 2/3 + 1/6 and have students solve it on whiteboards while you scan. Then one more with a twist (unlike denominators that aren't multiples). Only after most boards are right do students start the practice set.

• Modeling a written responseGrades 4–12

  Before students write an answer to a reading question, write one in front of them, thinking aloud: "The question asks why. So I need a reason and evidence. Let me find the sentence that shows it." Then write a second answer together with the class, and only then have them write their own.

• Five-minute daily reviewGrades 2–12

  Every lesson opens with five quick questions: two from yesterday, two from last week, one from last month. Students answer on paper, you go through answers, and you note which item most missed. Rosenshine's first and last principles (begin with review; review weekly and monthly) in five minutes.

• Small-group math interventionGrades 1–5

  A group of four students who struggle with place value meet with you for twenty minutes, four days a week. Each session: a short review, a clear model with base-ten blocks, guided practice with every student answering, then a few problems alone, then a quick timed fluency activity. It follows the math intervention guide's recommendations: systematic instruction, concrete representations and regular timed fluency practice.

Common mistakes

Talking too long.

Explicit instruction is short explanations and lots of practice. If you've been explaining for ten minutes without students doing anything, break the explanation into smaller steps with a question or a problem after each one.

Skipping "we do".

Going straight from your model to independent work leaves students who half-understood practicing errors. Guided practice, with every student responding, is where you find out who's ready.

Checking only volunteers.

Asking "Any questions?" or taking the first hand tells you about the confident few. Use all-student responses and random calling, as Rosenshine's principles describe.

Using it for everything, forever.

Guidance matters most for novices. As students gain knowledge, fade the worked examples and give more open problems; Kirschner and colleagues note the advantage of guidance shrinks as learners know more.

Confusing it with a script.

Scripted Direct Instruction programs are one well-researched option, but explicit teaching is a set of moves you can use with any curriculum. You don't need a script to state a goal, model, guide and check.

No review.

A skill taught once and never revisited fades. Build daily, weekly and monthly review into the plan, so earlier material keeps coming back.

Time and cost

Time: no more class time than any other lesson; the change is how the time is used. Planning takes longer at first, because worked examples and good questions have to be written in advance.

Money: none for explicit teaching. Scripted Direct Instruction programs are commercial and come with training costs.

Training: moderate. Most teachers already explain and model; the hard parts are breaking content into small enough steps, checking every student, and having the discipline to go back to guided practice when checks show it's needed.

Debates and open questions

Explicit instruction or inquiry? This is one of the longest-running arguments in education. Kirschner, Sweller and Clark argued that minimal guidance doesn't work for novices; Cindy Hmelo-Silver and colleagues replied that inquiry and problem-based learning are highly scaffolded and shouldn't be lumped with unguided discovery. The Alfieri meta-analysis suggests both are partly right: unassisted discovery does worse than explicit teaching, and well-supported discovery does well.

What about the scripted programs? The Stockard meta-analysis is large and consistently positive, but it covers the Direct Instruction family specifically, not explicit teaching in general, and the authors found effects were larger with more exposure, which is a reason to judge them over a year rather than a few weeks. Some teachers dislike scripts; the research is about outcomes for students, not about whether teachers enjoy the materials, and both are worth weighing.

Does it kill motivation or creativity? Critics worry that teacher-led lessons make students passive. The Stockard meta-analysis found the scripted programs' effects on attitudes and feelings weren't statistically significant, rather than negative. A well-run explicit lesson is unusually active for students, with constant questions and practice; a poorly run one is a lecture.

Tools that do it

Free, on this site, nothing to sign up for. Class lists stay in your browser.

• Random name picker: check every student, not just volunteers: ask, give think time, then pick names.
• Math worksheet generator: independent practice sets that mix today's skill with earlier ones.
• Base-ten blocks: model place value step by step on the projector, then have students do it with you.
• Flashcard maker: daily and weekly review of facts and vocabulary from earlier lessons.
• Countdown timer: keep explanations short and practice moving: three minutes to model, five minutes of guided practice.
• Exit tickets: a final check that tells you who's ready for independent work tomorrow.

Questions teachers ask

What's the difference between direct instruction and explicit instruction?

Explicit instruction (sometimes called direct instruction with small letters) is a general way of teaching: clear goals, modeling, guided practice, checks and review. Direct Instruction with capitals is a family of scripted commercial programs built on the same ideas. The research on each is different, and this card covers both.

Is explicit instruction just lecturing?

No. A lecture is long teacher talk with little checking. Explicit instruction is short explanations broken into small steps, with frequent questions that every student answers and practice after each step. Students in a good explicit lesson do more work, not less.

Does it work for advanced students?

Guidance matters most when students are new to the material. Kirschner and colleagues note the advantage of strong guidance shrinks once learners have enough knowledge to guide themselves, so advanced students can move to independent and open-ended work sooner.

What is I do, we do, you do?

A short name for the release of responsibility in an explicit lesson. I do: the teacher models. We do: teacher and students practice together, with the teacher fading support. You do: students practice alone. The checks between stages decide when to move on.

Is there research on the scripted Direct Instruction programs?

A lot. Stockard and colleagues' 2018 meta-analysis pooled 328 studies over fifty years and found positive effects on academic outcomes across reading, math, language and spelling, which held up after the programs ended.

Can explicit instruction work with hands-on learning?

Yes. Modeling with manipulatives, then guided practice with the same materials, is explicit instruction. The math intervention guide rates using concrete and semi-concrete representations as Strong alongside systematic instruction.