The Review Session That Isn't a Re-Lecture: Designing End-of-Term Review From the Exam Blueprint

In the last week of term, many courses hold some kind of review. The format is familiar: the instructor stands at the front and walks through the semester's material in order, faster than the first time, pausing now and then to ask whether there are any questions. A few students ask. Most copy down anything that sounds like a hint about the exam. Everyone leaves feeling that something useful happened, and the exam results often suggest otherwise.

A re-lecture is a comfortable review for the instructor and a passive one for students. It gives them the feeling of recognizing the material, which is different from being able to use it. If the final asks students to apply, analyze, or decide — and the exam blueprint says what it asks — then the review should have them doing exactly that, while there's still someone in the room to correct them.

Start from the blueprint, not the syllabus

The syllabus lists what was covered. The blueprint lists what the exam will measure and how heavily. Design the review from the blueprint. Look at the cells with the most weight, especially those that ask for higher-order thinking, and ask a simple question: when did students last practice this kind of task, and how did they do? The cells where the answer is “not recently” or “not well” are where the review time belongs.

This is backward design applied to a single class session. The desired result is performance on the final. The evidence is the kinds of task the blueprint describes. The learning activity — the review — should be practice at producing that evidence.

Replace explanation with retrieval

Students tend to remember material better when they have to pull it out of memory than when they simply see it again. That's why a review built around questions does more than one built around slides. Open with a short set of problems or prompts drawn from the heaviest-weighted parts of the blueprint, and have students attempt them before any explanation. Low stakes, no grade, but real effort.

What students get wrong during this attempt is the most valuable information in the session. It tells you, and them, exactly where the gaps are while there's still time to address them. Walk through the answers after they've tried, and spend your explanation time on the items many students missed rather than on the ones they already know.

Match the practice to the exam's demand

If the final will ask students to analyze an unfamiliar case, the review should give them an unfamiliar case to analyze. If it will ask them to write an argument, they should draft one. Practice that matches the demand of the exam is more useful than practice that only touches the same content. A student who can define every term in the unit may still freeze when asked to use those terms to explain something new.

Don't recycle the exact problems from earlier in the term. Students who recognize a problem are recalling an answer, not practicing the skill. Change the context or the scenario so that the task is familiar but the answer isn't.

Plan the session's time the way the blueprint plans the exam's weight. If a cell carries a large share of the final, it should carry a similar share of the review. It's easy to spend half the session on the topic students ask about most, which is often the one they're most anxious about rather than the one that matters most. Answer those questions, briefly, and then bring the time back to what the exam will actually weigh.

Make students do the explaining

Good review activities put students in the position of explaining. Have them work a problem in pairs and then explain their reasoning to another pair. Ask them to write a short explanation of a common mistake and why it's wrong. Give them a worked solution with an error in it and ask them to find it. Each of these requires students to understand the material well enough to talk about it, and each shows you where their understanding breaks down.

Tell students what the exam looks like

Sharing the shape of the exam isn't giving it away. Tell students which outcomes carry the most weight, what kinds of tasks they'll face, and how their answers will be judged. If you're using a rubric for written responses, show it and have them apply it to a sample answer. Students who know what's being measured spend their study time on the right things, and the results then tell you more about what they learned than about what they guessed you'd ask.

Send them away with a study plan

End the session by having each student write down the areas where they struggled most during the review and one specific thing they'll do about each before the exam. It takes a few minutes, and it turns the review from something that happened to them into a plan they own. If you collect the lists, they also tell you which parts of the course need attention next term — the same information a first-exam postmortem gives you, but before the final rather than after.

The bottom line

A review session that re-lectures the course gives students the feeling of knowing without the practice of doing. Design it from the exam blueprint instead: find the heavily weighted, higher-order cells students haven't practiced recently, and give them retrieval practice that matches what the exam will ask. Let them attempt before you explain, and spend your time on what they missed. Put students in the position of explaining, share the shape of the exam and how it will be judged, and send them out with a plan of their own. The last class before the final should be the one where students do the most work.

TeachingsByDesign builds the exam blueprint from your course outcomes, tags every item to its cell, and flags the demand your course never gave students a chance to practice — so the review session goes where the final will. See how it works.