Practice testing vs rereading: why self-testing wins
Practice testing beats rereading because it makes you retrieve information instead of recognizing it, and retrieval is what strengthens memory. Rereading makes material feel familiar, which is easy to mistake for knowing it. Research on the testing effect shows that people who quiz themselves remember noticeably more days later than people who reread.
Ask a room of students how they study and “reread my notes” wins easily. Ask the research what works and rereading lands near the bottom. That gap between the most popular method and one of the weakest is one of the cheapest grade improvements available.
Why rereading feels like it works
When you reread a chapter, it looks familiar. You nod along. “Yeah, I know this.” But recognizing a concept when it’s printed in front of you is a different skill from producing it on a blank exam page, and exams almost always ask for the second one.
Picture an intro psych student rereading the section on classical conditioning. Unconditioned stimulus, conditioned response, extinction: every term looks friendly. Then the exam describes a cat who now sprints to the kitchen whenever the can opener whirs and asks them to label each part, and the terms won’t come. The rereading built recognition. It didn’t build recall.
The trap is that rereading is pleasant because it’s easy, and it’s easy because you aren’t retrieving anything. That comfort is the warning sign.
What the research says about practice testing
The testing effect is one of the most replicated results in learning science. A classic demonstration comes from Henry Roediger and Jeffrey Karpicke in 2006. Students read short prose passages and then either restudied them or took recall tests on them. On a test five minutes later, the restudy group did a little better. A week later, the pattern flipped: the students who had tested themselves remembered considerably more.
That flip matters. Rereading wins in the short term, which is exactly why it feels effective the night you do it. Testing wins on the timescale your exam actually happens on.
In 2013, John Dunlosky and colleagues reviewed ten popular study techniques and rated them by how well the evidence supported them. Practice testing and spacing out study sessions came out at the top as “high utility.” Rereading and highlighting were rated “low utility.”
A practice test also does something rereading never will. It hands you a list of what you don’t know. Find those gaps a week before the exam and you can fix them. Find them during the exam and, well.
How to take practice tests that actually help
Not every quiz counts. These are the habits that separate useful practice testing from a ritual.
Make it closed-book and timed
A test with your notes open is rereading in disguise, because the retrieval never happens. Put the phone in another room, set a timer that matches the real exam, and answer from memory. If you blank, write down what you can and move on, the way you would in the exam hall.
Grade honestly, then study only the misses
The score is the least interesting part. The list of wrong answers is your next study session. Going back over questions you nailed feels nice and teaches you almost nothing. For each miss, figure out whether you didn’t know it, half-knew it, or misread the question, because each needs a different fix.
Practice the hard question types
If your organic chemistry exam has mechanism problems, a deck of multiple-choice definitions won’t prepare you. If your history survey final has an essay, write an outline under time pressure. Practice in the format that costs the most points.
Space your tests out
One test the night before is cramming with extra steps. Two or three spread across a week or two let the forgetting happen in between, and retrieving something you’ve half-forgotten is what makes it stick. The mechanics of that are in spaced repetition without the overwhelm.
Where to get practice tests for your class
The best practice test looks like the real one. In rough order of usefulness:
- Old exams and quizzes from your professor. Many post them; others will share if you ask in office hours. Question styles get reused constantly.
- End-of-chapter problems, especially in STEM. In Calc II or thermodynamics, working problems cold is the practice test.
- Your own questions. Turn each heading and learning objective in the syllabus into a question. If you took Cornell notes, the cue column already is one.
- A study partner’s questions. Swap ten questions each. Writing them forces you to think about what’s likely to be tested, which is useful on its own.
Writing good questions takes time, and that’s the real reason most students fall back on rereading. It’s also exactly the kind of setup work worth handing off.
How MyStudySidekick makes the practice tests for you
Import your lectures, slides, and readings, and MyStudySidekick generates quizzes and full mock exams from your own material, including multiple choice, short answer, and open-ended questions. When you finish, it grades the test, written answers included, and gives feedback on each miss so you know what to restudy. The most effective study method stops being the most annoying one to set up.
Common practice-testing mistakes
- Peeking. If you can look, you don’t retrieve, and retrieval is the whole mechanism.
- Chasing the score. A 90 on easy questions tells you less than a 60 on hard ones.
- Using only multiple choice. It’s easier to recognize the right answer than to produce it. If the exam asks you to produce, practice producing.
- Testing once, the night before. There’s no time left to fix what it reveals.
A two-week practice testing plan
If your exam is about two weeks out, swap half your rereading time for this:
- Day 1: take a short closed-book quiz on everything covered so far. Don’t prepare first; you want an honest baseline.
- Days 2 to 5: study only the topics you missed, using active recall rather than rereading. Active recall techniques has specific ways to do it.
- Day 6: a second quiz on the same topics, reshuffled. Check whether the gaps closed.
- Days 7 to 11: fold in newer material and keep quizzing in short sessions.
- Day 12 or 13: a full-length, timed mock exam in the real format. The misses become your final study list.
- Day 14: light review and a full night of sleep.
You’ll walk in knowing what you know instead of hoping. If you’re working with less time than that, preparing for finals in a week compresses the same idea.
Rather not write the questions yourself? Let MyStudySidekick build your first practice test.
Frequently asked questions
Is rereading an effective way to study?
Not very. In a 2013 review of ten common study techniques, Dunlosky and colleagues rated rereading as low utility because its benefits are small and fade quickly. It can help on a first pass to understand material, but repeating it makes content feel familiar without making it easier to recall. Self-testing is a much better use of the same time.
What is the testing effect?
The testing effect is the finding that retrieving information from memory strengthens it more than restudying the same information. In a well-known 2006 experiment, Roediger and Karpicke found that students who took a recall test on a passage remembered more of it a week later than students who spent that time rereading it, even though rereaders did slightly better on an immediate test.
How many practice tests should I take before an exam?
Two or three spread across the final week or two is a good target for most courses. The first test shows you what you do not know, a second a few days later checks whether you fixed it, and a final timed one close to the exam builds familiarity with the format. More is fine if you restudy your misses in between.
Where can I find practice tests for my class?
Start with old exams and quizzes from your professor or course website, end-of-chapter problems in the textbook, and question banks that come with the book. You can also write your own by turning headings and learning objectives into questions. Study groups can swap questions, which forces everyone to think about what is likely to be tested.