Spaced Repetition for College Students: A Practical Setup for Midterms and Finals

Spaced repetition means reviewing the same material several times with gaps in between, and making each review a test: you try to answer before you look. It's mid-October, so you probably have a midterm this week or next, and if your finals fall in early to mid-December, they're about eight weeks away. That leaves enough time for cards you write today to come back four or five times before the final.

Most guides stop at "space out your studying." This one covers the setup: how far apart to put reviews for the exam you actually have, how to write cards that hold up, how many new cards a day you can keep up with, and whether paper boxes or an app scheduler is the better choice. If you want the general research on recall and spacing first, our midterm guide has it.

How spaced repetition works, and why it's more than starting early

Two things have to happen at once. The reviews have to be spread out, and each one has to be retrieval: pulling the answer out of memory, not rereading it. A 2021 meta-analysis of 29 studies on exactly this combination found a large advantage for spaced retrieval over the same number of retrievals done back to back (Hedges' g = 0.74, after correcting for likely publication bias) (Latimier, Peyre & Ramus). The big 2013 review of study techniques by Dunlosky and colleagues gave both ingredients, distributed practice and practice testing, its highest utility rating (Dunlosky et al., 2013).

The reason spacing matters is the shape of forgetting. In 2015, two researchers repeated Hermann Ebbinghaus's 1880s self-experiment: one person spent 70 hours learning lists and relearning them after delays from 20 minutes to 31 days, and the curve came out much like the original, steep early and flatter later (Murre & Dros, 2015). A review placed on that flatter part, when the answer takes some effort to find, does more than one placed while it's still fresh. In the free forgetting-curve simulator, each review you add resets the curve and slows the next drop, and passive review moves it far less than the more effortful methods. For the psychology of why effortful recall sticks, see the free AP Psychology sections on retrieving memories and forgetting, or the Learning and Memory section of the free Psychology as Science course.

How far apart to space reviews: match the gap to the exam

There is no single best interval. The best gap depends on how long you need to remember the material, and the clearest data on this comes from a study of more than 1,350 people, recruited online, who learned 32 obscure trivia facts, reviewed them once after a gap of up to 3.5 months, and took a final test up to a year later (Cepeda et al., 2008). Of the gaps the researchers tried, these gave the best recall:

Two practical rules come out of that table. First, the longer you need something, the longer you can wait between reviews, although the gap shrinks as a share of the delay (from about 20 to 40% of a one-week delay to 5 to 10% of a one-year delay). Second, the penalty is lopsided. In the authors' manuscript of the paper, recall climbed steeply as the gap grew toward the best value and then fell off only gradually past it. Waiting too long costs you a little. Reviewing too soon costs you much more.

An earlier review of 317 experiments points the same way: the gap that produced the best retention grew as the time until the final test grew (Cepeda et al., 2006).

A spaced repetition schedule for this semester

Many study guides hand you one fixed schedule, such as days 1, 3, 7 and 14. That's a reasonable shape, but the Cepeda results say the right spacing depends on when your exam is. Those studies also used one review. Flashcards give you several, so here is a setup built on the same logic. Day 0 is the day you first learn or make the card.

Treat the numbers as a starting point. A card you missed goes back to a short gap, and a card you get instantly every time can stretch further. Material from the first weeks of the term, which your final will test, belongs on the 8-week row right now.

How to write flashcards that hold up

Card quality decides whether the reviews are worth doing. Anki's manual says the most efficient fix for a card you keep forgetting is to change how the information is presented, often because the card holds too much or you're memorizing something you don't understand yet (Anki Manual: Leeches). One research result is worth knowing before you download a shared deck: across six experiments, people who made their own flashcards from a text remembered more on a test 48 hours later than people who studied premade cards, and usually did better on application questions too (Pan et al., 2023).

The card-writing checklist

• One idea per card. "What are the three stages of memory?" is three cards, or one card plus a card for each stage.
• A question on the front, not a topic. "Mitosis" invites a vague nod. "What happens in anaphase?" forces an answer.
• Your own words on the back. Copying a textbook sentence makes cards fast to write and easy to recognize without understanding. Rewording forces the deeper processing described in the free AP Psychology section on encoding memories.
• A "why" card for each key fact. "Why does a variable-ratio schedule resist extinction?" Dunlosky's review rated this kind of self-explanation (elaborative interrogation) moderate utility.
• Both directions when the exam goes both ways. Term to definition and definition to term, or structure to function and back.
• Cover the labels on diagrams. For anatomy or cell biology, the front is the image with one label hidden.
• No cards for what you don't understand yet. Go to office hours or work an example first, then write the card.
• Answer out loud or on paper before you flip. "I knew that" after seeing the answer doesn't count as a correct recall.

Your lecture notes are the raw material: our guide to taking notes in college shows how to turn each page into questions as you go.

How many new cards a day

This is the part most students get wrong, and the cost shows up two weeks later. Every new card brings a run of reviews with it, because new material has to be seen several times before the gaps can grow. Anki's own manual puts a number on it: learning about 20 new cards a day, consistently, leads to roughly 200 reviews a day, and it warns that people who add hundreds of cards in their first few days end up buried in reviews (Anki Manual: Deck Options).

Work it out from your exams instead of picking a number:

1. Count what has to go on cards. Say a biology final covers about 400 terms and processes.
2. Decide when the last new card goes in. Anki's FAQ on preparing for a large exam suggests leaving 2 to 4 weeks between your last new material and the test, and planning for a few days you'll miss. For a final on December 11, stopping 3 weeks out means November 20, about 35 days from today; take off 3 missed days and you have 32.
3. Divide. 400 cards over 32 days is about 12 or 13 new cards a day, which by Anki's ratio means on the order of 100 to 150 reviews a day once the deck is running.
4. Check that against your week. If that's more than you'll actually do across all your classes, cut the card list, not the reviews.

How many times should each card come back? Three experiments with 533 students give a concrete target for concepts: practice until you've recalled each one correctly 3 times in the first session, then relearn it in 3 more sessions at widely spaced intervals. Relearning added a lot of long-term retention for very little extra practice (Rawson & Dunlosky, 2011). And don't retire a card the first time you get it right. In a foreign-language vocabulary study of university students published in Science, dropping words from testing once they had been recalled once hurt recall on a delayed test, while extra restudy did nothing, (Karpicke & Roediger, 2008).

To fit the daily round into a full week of classes, put it in a fixed slot, as in our guide to making a study schedule.

Leitner boxes vs an app algorithm

Leitner boxes: spaced repetition with physical flashcards

Anki's documentation credits Sebastian Leitner with popularizing paper-card boxes in 1972: cards move between boxes after each right or wrong answer, so you can see at a glance which cards are due (Anki Manual: Background). In the five-box version below, which fits the schedule table above, a correct card moves up one box and a missed card goes back to box 1:

Boxes cost nothing and work without a phone. Anki's own docs name the downside: boxes can't tell you an exact review date, and they handle cards of very different difficulty poorly.

App schedulers: what Anki's SM-2 and FSRS do

An app does the bookkeeping per card. Anki's legacy scheduler is based on SuperMemo's SM-2 algorithm: with the default starting ease of 2.50, answering Good on a card stretches the next delay about 2.5 times, so a 10-day gap becomes about 25 days. Its newer option, FSRS (Free Spaced Repetition Scheduler), which you turn on in the deck options, estimates each card's stability, defined as the number of days it takes for your chance of recalling it to fall from 100% to 90%, and schedules reviews for a target you choose (Anki FAQ: What spaced repetition algorithm does Anki use?). The default target is 90% retention. Anki's manual says the workload rises very quickly above 90% and can become overwhelming above 97%.

Which one should you use?

Less hangs on this than app marketing suggests. In the meta-analysis above, expanding gaps (the shape both boxes and apps use) and equal gaps came out almost even (g = 0.032, not significant). Expanding gaps showed a small edge only when each item was seen more than four times, and even that wasn't statistically significant. An earlier experiment found equal gaps did better two days later, and pointed to the delay before the first test as the factor that mattered (Karpicke & Roediger, 2007). Choose the system you'll open every day. Use an app if you have hundreds of cards or several classes; use boxes if a phone is where your study sessions go to die.

One adjustment either way: a scheduler aims at long-term memory, not at your exam date, so a card's next review can land after the final. Do one full pass of the deck in the last week regardless of what's due; in Anki, the custom study option lets you review cards that aren't due yet.

Mistakes that quietly break a spaced repetition system

• Pressing Hard when you forgot. Anki's manual says FSRS treats Hard as a correct answer, so using it for a miss makes every interval too long. A miss is Again.
• Skipping a week, then facing the pile. Set a maximum number of reviews a day. Anki's manual describes that limit as a way to smooth out peaks after a break; clear the backlog over several days instead of quitting.
• Keeping cards you fail over and over. Anki tags a card as a leech after 8 lapses by default and suspends it. Rewrite it, split it or drop it.
• Relying on cards for problem-solving classes. Cards suit definitions, formulas and when to use each one. For calculus, chemistry or physics, space whole worked problems too: every week or two, redo a mixed set from an earlier unit, such as AP Practice 1 on limits in the free AP Calculus AB course.
• Trusting how a session feels. In the Science vocabulary study above, students' predictions of how much they'd remember had no relationship to how much they did. Judge by your recall on cards you haven't seen in a week.

Start this week: a 15-minute setup

1. List your exams through December with their dates.
2. Start one deck per course, beginning with the course that has the most to memorize: term-heavy classes like biology and psychology are the natural first decks.
3. Write cards from this week's lectures only, using the checklist. Don't try to card the whole semester tonight.
4. Set your new-card number from the arithmetic above and a review cap you'll actually hit.
5. Do the first round tomorrow, at the same time you'll do it every day.

If an exam is tomorrow, spaced repetition can't help much for that one; use our one-day exam plan and start the deck afterward for the final. For ready-made material to practice with, Lykke's free courses on Explore, such as AP Biology and Anatomy and Physiology, come with flashcards and quizzes, and How it works shows the rest of what Lykke does.