Reteaching strategies: whole class, small group, or one student

Start from how the mistakes are spread

Whole class: one shared misconception

If half the class or more chose the same wrong answer, the first explanation did not land. Repeating it louder rarely works, because students who hold a misconception tend to fit the explanation into it. Change the representation instead.

Five ways to reteach differently

• A counterexample: show a case where the wrong idea gives an answer that is obviously false.
• A different representation: a timeline instead of a list of dates, a diagram instead of a definition.
• A worked example, then a near-identical one for students to finish themselves.
• A student explains the wrong method and why it seemed right; the class finds where it breaks.
• Sorting: give correct and incorrect examples and ask students to sort them and say why.

A worked example from a reading lesson

A Year 8 English class reads a short story and answers six questions. On the inference question, why the narrator hides the letter, 15 of 28 students choose the option that repeats a phrase from the last paragraph. That option sounds right because the words are familiar, but it does not answer the question.

That is a shared pattern, so it is a whole-class reteach, and repeating "read carefully" will not fix it. Instead the teacher puts the paragraph on the board, covers the answer options, and asks students to say in their own words what the narrator is afraid of before looking at the options again. Then two sorted examples: answers that repeat the text and answers that explain it.

A week later, a different story, a similar inference question: 23 of 28 get it right. The five who did not become a small group. The numbers in this example are illustrative.

A real example: a true fact that is not the answer

In Gaplight's history course, asked which literary work was written under Yaroslav the Wise, 37% of 54 answers chose Ruska Pravda instead of the Sermon on Law and Grace. Ruska Pravda does date from Yaroslav's reign, but it is a law code, not a literary work. The students knew the period; they missed the category.

A second question shows the same pattern: asked for the historical significance of Prince Oleg's rule, 34% of 53 answers chose his first treaty with Byzantium instead of the completion of the Kyivan state. The treaty is a real event from his rule, just not its main significance.

Repeating the facts would not help, because students already know them. What helps is sorting: give a mixed list of documents and events and ask students to sort them into law codes, chronicles and literary works, or into event and significance, and say why. The sorting is the reteach.

Real data: Gaplight's history course for Ukraine's national exam, snapshot of 7 October 2026. Counts are answers, not students: one student can answer a question more than once.

Small group: one skill, a few students

Four to eight students who miss the same skill can be taught together for ten minutes while the rest of the class practises. Group by skill, not by overall score: two students at 60% may need completely different things. Name the group by the skill ("the evidence group"), not by level, and dissolve it once its students get the skill right on new questions.

One student: many gaps

A student wrong across several skills is often missing something earlier in the course that the later skills depend on. Look for the earliest skill they get wrong and start there. In a history course, a student who mixes up causes, consequences and sources may simply be unsure of the order of events; fixing chronology can clear the rest.

Check that it worked

Reteaching is only done when the students get the skill right on questions they have not practised, a week or so later. Without that check it is easy to reteach the same thing twice, or never notice that it stuck. Short, regular reviews of earlier material, Rosenshine's first principle, give you that check for free.

Common mistakes

• Reteaching the topic rather than the specific wrong idea students hold.
• Using the same explanation, only slower.
• Grouping by overall score instead of by the skill missed.
• Keeping groups fixed for weeks, so students labelled as weak stay weak.
• Declaring success because students can now answer the question that was retaught.

Finding the groups without extra work

Gaplight reads every answer on a school's own checks and gives each teacher students grouped by the skill they miss, so the decision in the table above takes a minute rather than an evening.