How many questions should an end-of-unit review game have?

Twenty to twenty-five for a full period. That's enough to touch every objective with a difficulty ladder in each, and it fits in roughly thirty minutes of play with time left to go over what the class missed.

Should every category have the same number of clues?

Not necessarily. Category depth ideally tracks how much instructional time and assessment weight each objective carries. If your board tool requires a uniform grid, you can weight the unit through difficulty instead, giving the major objective the full ladder and keeping smaller categories at the lower end.

When in the unit should the review game run?

The second-to-last day, not the day before the test. Running it a day earlier leaves you a working day to reteach whatever the game exposed. A review on test eve identifies gaps at the point you can no longer close them.

Where should the questions come from?

Your own unit materials, especially exit tickets students missed, homework items that generated repeated questions, and anything you already had to reteach once. Those are worth more review time than freshly written questions, because you already have evidence students struggled with them.

How do I know what to reteach after the game?

Look for the column with the most misses, the row where a category's ladder broke, and any clue where several teams gave the same wrong answer. Shared wrong answers often point to a single misconception, which is an efficient thing to fix in one lesson.

End of Unit Review Games: Building Boards From Your Objectives

Many end-of-unit review games are built the night before from whatever the teacher happens to remember teaching. The result is predictable: heavy coverage of the last week, thin coverage of the first, and no coverage at all of the two objectives that were taught quickly and are about to be tested.

Building from the unit plan instead takes about the same amount of time and produces a genuinely different session.

Start from the objectives, not the questions

Open whatever document actually defines the unit: your objectives list, the standards you mapped, or the assessment blueprint if you have one. That document is your category list.

A five-category board maps cleanly onto a unit with five objectives. If your unit has seven, you have three options:

Combine two closely related objectives into one category
Use six categories and accept a slightly wider board
Leave out the objective that carries the least assessment weight, and say so to the class

The third option is legitimate as long as it's a decision rather than an accident.

Weight the board by instructional time

If you spent six days on one objective and one day on another, a board that gives them equal space misrepresents the unit.

A workable rule: the number of clues per category should roughly track the number of lessons you spent on it. That might mean:

Major objective taught over a week: 7 clues
Two mid-sized objectives: 5 clues each
Two smaller objectives: 4 clues each

Not every board tool allows columns of different depths, so check what yours supports before planning around it. If you're working with a uniform grid, you can weight the unit a second way: give the major objective's column the full difficulty ladder from recall up to synthesis, and keep the smaller categories concentrated at the lower end.

Build a real difficulty ladder inside each category

Within a column, the point values should mean something. A useful progression for most subjects:

Recall. Define, name, identify.
Recall with context. Identify within an example.
Application. Use the concept on a new case.
Multi-step. Two or more operations combined.
Synthesis or comparison. Connect this objective to another one in the unit.

That top row matters more than it looks. Clues that connect two objectives are the ones that reveal whether students have a unit or a pile of disconnected lessons.

Pull questions from your own materials

The fastest source of good review questions is material you already made:

Exit tickets from the unit, especially ones a lot of students missed
Practice problems that generated questions in class
Homework items students asked about repeatedly
Anything you had to reteach mid-unit

Questions students already struggled with are usually worth more review time than fresh ones. If you kept exit tickets, the ones with the worst results are your high-value clues.

If you're generating questions from your notes or slides rather than writing them by hand, verify each one against what you actually taught before it goes on the board. Generated questions can be plausible and still cover something you skipped.

A sample board

An eighth-grade science unit on energy, built this way:

| Category | Clues | Why | |---|---|---| | Forms of Energy | 5 | Foundational, taught over three days | | Energy Transfer | 7 | The unit's core objective, a full week | | Conservation of Energy | 5 | Three days, heavily assessed | | Energy in Systems | 4 | Two days | | Calculations | 4 | Two days, but high test weight |

The Energy Transfer column runs deepest because that's where the instruction went. The Calculations column is short but the top two clues are multi-step, because that's where the assessment pressure sits.

Timing the review within the unit

Run the review game the second-to-last day of the unit, not the last.

This is a common scheduling mistake. A review on the day before the test tells you what students don't know at the exact moment you can no longer do anything about it. Running it one day earlier leaves a working day to reteach whatever the board exposed.

A workable close to a unit:

Day A: Review game. Identify gaps.
Day B: Reteach the two weakest areas, plus independent practice.
Day C: Assessment.

If the unit calendar only allows two days, run the game on the first of them and keep the reteaching short and targeted.

Reading the results

The review is diagnostic, so treat the output as data rather than as a scoreboard.

What to note during play:

Which column produced the most misses. That's your reteach target.
Where the ladder broke. If teams cleared rows one through three in a category and collapsed at row four, students have the recall but not the application. That's a different reteach than not knowing the content at all.
Clues where multiple teams gave the same wrong answer. A shared wrong answer often indicates a shared misconception, which is more fixable than scattered confusion.
Clues nobody attempted. Teams tend to avoid categories they feel weak in. An untouched column is a signal.

If every student wrote answers during play, the collected papers give you a per-student version of the same picture. That method is described in [how to play a Jeopardy-style game in the classroom](/articles/how-to-play-jeopardy-in-the-classroom).

Reusing the board

The reason to build from objectives rather than from memory is that the structure survives. Next year the categories are the same because the unit is the same. You rewrite clues, swap in new examples, and adjust anything the class found ambiguous.

The first build takes a full planning period. The second takes twenty minutes.

A board built this way is also the honest starting point for the pre-test review, since it already maps to the unit's objectives.

For building the board itself, Kleveroo's [Jeopardy board maker](/jeopardy-board-maker) is one option, and you can browse existing boards for structure ideas before creating your own. Alternatives to the board format are covered in [classroom review games for any subject](/articles/best-classroom-review-games).

End-of-Unit Review Games That Actually Cover the Unit