Back to blog

Open Book Exams: Do You Still Need Flashcards?

Yes, but far fewer, and different ones. In an open book exam the scarce thing is time, so you card the map of your notes and the facts you need at speed.

Yes, you still need some flashcards for an open book exam, but a lot fewer than for a closed book one, and they should be different cards. You are not carding the content anymore. You are carding where the content is, plus the small set of things you will need so often that looking them up every time would eat the exam.

The reason is simple and it catches people every term. In an open book exam, the scarce thing is not information. Everybody in the room has the same information sitting on the desk. The scarce thing is minutes. The exam is written knowing you have the book, so the questions are longer and harder, and the people who run out of time are almost always the ones who planned to look everything up.

For transparency, I work on Nebulearn, a flashcard app. This is one of those topics where the right answer is "fewer cards," and I would rather tell you that than sell you a bigger deck.

The lookup math nobody does beforehand

It is worth doing the arithmetic once, because it changes how you prepare.

Say the exam is two hours and has eight questions, so fifteen minutes each. Now think about what one lookup actually costs. You read the question, you think, "I know that is in chapter six somewhere," you find chapter six, you flip a few pages, you find the section, you read enough of it to check it is the right thing, and you go back to your answer and remember where you were. That is a minute if you are quick and organized, and two or three if you are not.

Now say each question needs you to look up four things. A formula, a definition, a table value, and a worked example to check your setup. At a minute and a half each, that is six minutes out of your fifteen gone on finding things. Across the exam it is forty eight minutes, which is forty percent of your time spent not answering questions.

That is the cost for somebody who knows where everything is. For somebody who shows up thinking, "it is open book, I will figure it out," every lookup starts with a search, and the search itself can take five minutes, and somewhere around question five they start writing answers from memory anyway, except they never built the memory.

So preparation for an open book exam has two jobs. Cut the number of lookups, and make each remaining lookup fast. Flashcards help with the first. A good index helps with the second.

Three piles: card it, tab it, leave it

Go through the course material and sort everything into three piles. This sorting is most of the studying, honestly, because to decide which pile something goes in, you have to understand what it is for.

PileWhat goes in itWhy
Card itCore definitions, the vocabulary the questions are written in, the formulas you use in nearly every problem, and the rules for choosing between methodsYou will need these many times per question. A lookup each time is too slow, and you cannot even read the question without them.
Tab itLong tables, rarely used formulas, constants, worked examples, lists of exceptions, anything with a lot of detailYou need to find it in ten seconds. You do not need it in your head.
Leave itDerivations you will not be asked to reproduce, background reading, anything the professor said is not examinableIt is not worth a card and it is not worth a tab.

The first pile is small. For most courses it is somewhere in the range of thirty to sixty cards, not three hundred. That is the whole point of the format: the exam is taking the memorization burden off you, so take it up on the offer for everything in the second pile.

The mistake people make is with the first row, in both directions. Some people card nothing, because "it is open book," and then they cannot parse the questions fast enough. Other people card everything out of habit, spend three weeks on a closed book style deck, and never build the index. Then they sit in the exam with a head full of table values they could have looked up and no idea which page the tricky example is on.

Card the map, not the territory

Here is the kind of card that only makes sense for an open book exam, and it is the one I would add first.

A normal card asks, "what is X?" A map card asks, "where is X, and when would I go there?" Like this:

Front: A question gives you a beam with a distributed load that is not uniform. Where in your notes is the method for that, and what is it called?

Back: Week 5 notes, the section on equivalent point loads, about three pages in. Tabbed in green. Replace the distribution with one force at its centroid.

You are not memorizing the method. You are memorizing that the method exists, what it is called, what kind of question triggers it, and where it lives. That is exactly what you need in the first thirty seconds of a question, when you are deciding what to do. Once you know you are going to the green tab in week five, the lookup takes ten seconds instead of two minutes.

There is a second kind of map card, which is a "which one" card. Open book exams love questions where there are three similar methods and the skill being tested is picking the right one. The book will happily tell you how to do each method. It will not tell you which one this question wants. So you card the decision:

Front: Two sample means, same subjects measured before and after. Which test, and why not the ordinary two sample one?

Back: Paired t-test. The measurements are not independent, and the ordinary two sample test assumes they are.

That card is worth a great deal in an exam where the formula for both tests is sitting open in front of you. It is the same idea as the recognition cards in flashcards vs practice problems: the choice of tool is a fact, and facts go on cards.

The index is the other half

The cards cut down how often you look things up. The index makes the remaining lookups fast, and building it is the best studying you will do for this kind of exam.

An index here just means one or two pages, written by you, that lists every topic and where to find it. Topic on the left, location on the right, plus a few words on when you would use it. If the exam allows your own notes, this goes on top of the pile. If it only allows the textbook, you do the same thing with sticky tabs and a colour code, and you memorize the colour code with three or four cards.

Some things that make an index actually work under pressure:

  1. Organize by question type, not by chapter. The exam does not say "this is a chapter six question." It says "find the deflection." Your index should have a line that says "deflection," with every place that word matters.
  2. Write it by hand, or at least write it yourself. Somebody else's index is a list of words. Yours is a record of you having gone through the material and decided what each piece is for.
  3. Put worked examples in it. "Example of a two-stage problem with losses: page 214." Under time pressure, a worked example you can pattern from is worth more than the theory section.
  4. Test it. Take an old exam question, start a timer, and see how long it takes you to find what you need. If it is more than twenty seconds, the index needs fixing, and you found that out on a Tuesday night instead of in the exam.

That last one is the step people skip, and it is the one that turns the index from a nice document into a tool. If you are short on days, the three day audit I wrote for high stakes decks adapts well here: the same idea of testing your material against real questions before the exam does it for you.

The one page aid sheet is its own case

A lot of exams are not fully open book. They let you bring one sheet, sometimes one side, sometimes both, sometimes handwritten only. That is a different game and it is worth treating separately.

With one sheet, space is the scarce thing, and deciding what goes on the sheet is the studying. Every line you write is a line you decided you could not trust yourself to remember, and every line you leave off is a bet that you have it. So the process is: make the cards first, review them for a week or two, and then build the sheet out of whatever you are still getting wrong. The things you have nailed do not need the space. The things that keep slipping do.

That gives you a nice loop. Your failed cards tell you what goes on the sheet. It is the leech list put to a different use: instead of deleting the cards you keep failing, you promote them to the aid sheet and stop worrying about them.

A sheet you copied from a friend the night before is close to worthless, for the same reason using someone else's flashcards tends to be. You do not know where anything is on it, you do not know what their abbreviations mean, and you spend the exam reading it like a foreign document.

When cards are not worth it at all

There are versions of "open book" where I would not make a single card.

If it is a take-home exam with several days, time is not scarce, and the right preparation is to understand the material well and have your sources organized. Cards are for retrieval at speed, and you do not need speed.

If the exam is open book and open internet with essay questions, what is being tested is your argument, and you would be better served by outlining answers to likely prompts. I wrote about that in how to study for an essay exam.

And if the course is almost entirely problem solving with a formula sheet provided, the formula sheet is the book, and your time should go to problems. The small deck of conditions and recognition cues is still worth having, but it might be twenty cards.

On tooling, if the only thing you need is a searchable copy of your notes, a PDF reader with a good search function is a better tool for this than any flashcard app, including mine, when the exam allows a laptop. Being able to search a word across four hundred pages in two seconds beats any tab system. Check the rules first, because many open book exams are paper only, and plenty of students find that out too late.

When this goes wrong on the day

Even with a good index and a small deck, a couple of things go wrong in the room.

The first is the confidence lookup. You know the answer. You are fairly sure. But the book is right there, so you think, "let me just check," and that is ninety seconds. Then you do it again on the next part. Checking things you already know is the biggest time leak in an open book exam, and the only fix is deciding beforehand that you will check at the end, if there is time, and not as you go.

The second is discovering during the exam that you do not understand something, and trying to learn it from the book right there. That almost never works. A textbook section that took you forty minutes to understand at home is not going to make sense in four minutes under pressure. If you hit one of those, write what you can, move on, and come back.

Both of those come from the same misunderstanding, which is treating the book as a safety net. It is closer to a reference shelf in a workshop: useful to someone who already knows what they are reaching for. The split on this page is that small core you carry in your head against the large body you only need to be able to find. That is how I would set it up in Nebulearn too: a short deck of definitions, decision rules and map cards, with FSRS on by default so thirty cards take a few minutes a day, and the rest of your prep time going into the index and old exams.

Open Book ExamsFlashcardsExam PrepStudy Workflow