The testing effect: why quizzing beats rereading

Ask students how they prepare for a test and most will describe the same routine: reread the notes, highlight the important parts, reread again. It feels productive — the material looks more and more familiar. And it's one of the least effective ways to remember anything.

What works better is less comfortable: closing the book and trying to recall. This guide explains the testing effect — what it is, the research behind it, how it differs from the spacing effect, which kinds of questions trigger it, and seven ways to build it into lessons and training sessions without adding a single graded test.

What is the testing effect?

The testing effect is the finding that retrieving information from memory — answering a question, recalling a definition, solving a problem without looking — makes that information easier to retrieve later, more so than spending the same time restudying it. It goes by several names: retrieval practice in education research, test-enhanced learning in psychology papers, and active recall among students.

The key distinction is between recognition (the answer is in front of you and looks familiar) and recall (you have to produce it). Rereading trains the first; testing trains the second, which is what exams and real work demand.

Rereading builds a feeling of fluency that is easily mistaken for learning — in the 2006 study below, the students who reread more were also more confident, and remembered less.

Who discovered the testing effect?

1917 — Recitation. Psychologist Arthur Gates found that students who spent part of their study time reciting the material from memory learned more than those who only read it.

1939 — 3,605 sixth graders. Herbert Spitzer tested pupils in 91 elementary schools across nine Iowa cities. His conclusion: "Immediate recall in the form of a test is an effective method of aiding the retention of learning" — and the benefit was still visible 63 days later.

2006 — Test-enhanced learning. Henry Roediger and Jeffrey Karpicke had students study short science texts. When the final test came five minutes later, restudying won. After two days or a week, testing won — and in the second experiment, the repeatedly tested group recalled 61% a week later against 40% for the group that only restudied, despite reading the text about 3.4 times instead of about 14.2.

2013 — Ten study techniques compared. A review led by John Dunlosky, published in Psychological Science in the Public Interest, rated ten common learning techniques. Practice testing and distributed practice received the highest utility rating; rereading, highlighting and summarizing were among those rated low.

How does the testing effect work?

Every act of retrieval is a small reconstruction — the brain has to search for the information and rebuild it, and that effort strengthens the path back to it. Psychologist Robert Bjork calls this kind of productive struggle a "desirable difficulty": learning that feels harder in the moment and lasts longer.

Two side effects matter for teachers: retrieval shows the gaps (a wrong answer reveals exactly what to reteach, for the student and for the teacher) and it supports transfer (ideas that have been retrieved are easier to apply in new situations).

What is the difference between the spacing effect and the testing effect?

The two effects change different parts of studying:

The two work best together — a recall question asked at spaced intervals is the simplest strong study routine there is.

What kinds of tests trigger the testing effect?

The common formats, from the most retrieval effort to the least:

Recall formats usually produce the strongest effect, but multiple-choice questions work too when the wrong options are plausible — a 2012 study by Jeri Little and colleagues found that competitive alternatives push students to retrieve information about each option, not just to recognize the right one. Two conditions make the difference: feedback right after answering (otherwise a chosen wrong option can stick) and distractors built on real misconceptions. Their advantage is practical: they are the easiest format to run often, with every student answering at once.

Seven ways to use retrieval practice in class

1. The three-question opener — Start the lesson with three questions on last week's topic, not yesterday's: retrieval after a gap is the strongest kind.
2. The mid-lesson check — One question after each key concept, answered by everyone at once; the answer distribution tells you whether to move on.
3. The exit ticket — One or two recall questions before students leave; you read the answers tonight and plan tomorrow.
4. The cumulative Friday quiz — Ten low-stakes questions mixing this week's content with topics from earlier units, so old material keeps coming back.
5. The brain dump — Two minutes, a blank page: "write everything you remember about today's lesson." Then compare with a partner.
6. Flashcards done right — Students say or write the answer before flipping. Cards can be generated from the lesson PDF with the free PDF to Flashcards tool.
7. The pretest — Five questions before a new unit. Students won't know most answers, and that's the point: trying first primes them for the lesson.

None of these needs a grade. The moment a retrieval check counts toward the report card, students start protecting their score instead of exposing their gaps.

The testing effect in corporate training

The same mechanism explains why so much training is forgotten within weeks — a slide deck is pure rereading.

Three habits change that: a knowledge check after every module (three to five questions, instant feedback); a follow-up quiz one week later and another a month later on the key points, so spacing and testing work together; and scenario questions instead of definition questions, so participants retrieve what to do, not just what something is called.

What are the limitations of the testing effect?

It needs something to retrieve. Quizzing on material that was never taught produces guessing, not learning — the pretest is the one exception, and it works only when teaching follows.

It needs feedback. Without the correct answer, errors can be reinforced, especially with multiple choice.

It needs low stakes. High-pressure tests add anxiety and push students back to cramming.

Its size varies. Effects are well established for facts, concepts and procedures; researchers still debate how large they are for very complex material.

None of these is a reason to avoid it — they are the instructions for using it well.

Running low-stakes quizzes with PanQuiz

PanQuiz makes the frequent quiz quick to build and quick to run. Questions can be generated by AI from a topic or from your own documents with the AI quiz generator, or from a PDF with the free PDF to Quiz tool, and every question stays editable. Students join with a PIN from any browser, with no account; results appear per question right after each one, so feedback is immediate; and question banks with random extraction make the cumulative Friday quiz different every week. Licenses are per teacher with education discounts; the 15-day Premium trial needs no credit card.

PanQuiz doesn't schedule spaced reviews for you — you decide when to bring a topic back; PanQuiz makes each quiz take minutes instead of an evening.

Four mistakes that cancel the testing effect

Grading every quiz. Retrieval practice is practice; graded, it becomes a test students prepare for by cramming.

Only recognition questions. If every question can be answered by spotting a familiar phrase, there's little retrieval — mix in recall and plausible distractors.

No feedback. Students need to see the right answer, and why, right away.

Once and done. One quiz helps; the same content retrieved again a week later helps far more.