What's the best way to study for a test?

When I was in school, teachers often implored us to not put off studying to the last minute. Sometimes they even suggested that we spread out our studying over a period of weeks. But who has time for that? Most of us just studied the night before the test -- with varying results, of course.

But surely research has been done on the ideal way to study. Is it possible to over-study? How much studying is enough? Wray Herbert has uncovered some real data on the problem, from a study by Doug Rohrer and Harold Paschler:

They had two groups of students study new vocabulary in different ways. One group drilled themselves five times; these students got a perfect score no more than once. The others kept drilling, for a total of ten trials; with this extra effort, the students had at least three perfect run-throughs. Then the psychologists quizzed all the students, once one week later and again three weeks after that.

After a week, the students who had done the most studying scored higher, but the advantage disappeared after three weeks. So cramming a whole lot right before the exam might help, but not if you study too early. But what if you take a study break? Rohrer and Pashler repeated the study with breaks varying from 5 minutes to a month. Here's what they found:

Those who had taken a one-day break performed best when they were tested ten days later. But if they were tested six months later (the laboratory equivalent of long-term learning), the optimal break time was a full month. In other words, as reported in the August issue of Current Directions in Psychological Science, "massing" all the study on a single topic together diminishes learning. It's better to leave it alone for a while and then return to it, and indeed the longer you want new learning to endure, the longer the optimal break between study sessions.

I wonder if this effect would be maintained over really long time periods: If I returned to studying Italian after 19 years since college, maybe I'd be a true master! Actually I tried that a couple years ago before I visited Italy. I'm pretty sure my mastery of the language didn't equal the level I was at in college, but I was surprised at how much I did remember.

Wray says Rohrer and Pashler's results even extend beyond memorization and into abstract learning like mathematics. Sounds like their advice isn't much different from what my teachers told me all those years ago. Now if only a researcher could explain how to get kids to listen to their teachers' advice!


More like this

I remember when I was studying for Step I of the medical Boards. Step I is the first of three very large tests that you have to take to become a doctor. This first test comprises everything you learn in the first two years of medical school, and it can in theory include the pathology and…
During my brief tenure as a high school teacher, one common suggestion I got from supportive colleagues was to "make your tests teaching tools." "That's often the only time you've really got your students' attention," they suggested, "so don't neglect the opportunity to teach them something." What…
While our exams were weeks ago, I know that some folks (especially high school students) are just finishing up. So these observations sent to me by a reader may be timely: I believe that if students are passing their classes with a B and above they should not have to take final exams. Most…
From today's Quotes Of The Day: As something of a student of history, I need to remember a number of numbers. Few of them remain easily in mind, although I normally can remember 1066, 1492, and 1776. It happens that on this day in 1776, Lorenzo Romano Amedeo Carlo Avogadro was born at Turin, Italy…

Cool! I've always noticed that effect, so glad the researchers are onto it now too. I've even found that sometimes I come back after a break ahead of where I left off, the brain can sort things out better sometimes if you leave it alone for a while.

These are interesting results, but it seems to me that there is enough of a distinction between rote memorization of lists of facts and the restructuring of knowledge that happens in "abstract" learning like mathematics. I've often found that, while I can't recall the specific theorem I need to answer an exam questions, I can often derive the result I need from what I know about the basic roadmap of whatever sort of math is being tested.

However, if I remember correctly, one of the take-home lessons from the rote-memorization-of-words literature is that people memorize words better the more they are able to put them into a meaningful context. To me, this suggests that rote and higher-order ("abstract") knowledge are related in important ways.

Dave, this site ROCKS!!! I'm teaching an intro to psych class, and I think this place is gonna start being required reading. Your little tidbits every day are so often SPOT ON.

That said, this particular article does a fine job of reinforcing what I have been saying as a teacher; study over longer periods of time. And like you said, where is the means of getting these critter to LISTEN to that sound advice?

Regarding your Italian after 17 years, there have indeed been just these sorts of super long-term retention tests, on foreign languages (Spanish) and mathematics. The name in this area is Henry Bahrick. Google scholar will turn up the pertinent papers.

By Shane Hoversten (not verified) on 28 Aug 2007 #permalink

iditya, that is fine. But might it be MORE fine if there is another strategy that worked better?

Besides, research in the post shows that it serves you fine for tests. It does NOT work for long-term retention, which is one of the ultimate goals you should (have) be(en) striving for as a student.

This seems to bear out the notion that spaced repetition is preferable to massed practice in second language vocabulary learning. Of course, as alluded to above, working memory differences might also mediate--perhaps in terms of how long the training sets should ideally last. Regarding attrition, frequency of exposure might be the biggest factor, but there could be others: perceptual salience (e.g. Japanese or Chinese characters having a great number of stokes), degree of L1 similarity (or, cognates), semantic information encoded by the word, etc. and one might also expect to see some differences on production vs. reception tests. Great work on this site!

(John) Anderson and Schooler (1991) is an excellent paper on this subject.

The people you really want to speak to and work with are musicians and teachers thereof. They have known forever that distributed practice outperforms massed practice. Moreover, since music involves both cognitive and motor functions (as opposed to, say, math) findings from this domain might be more easily generalized.

By Howard Weinberg (not verified) on 29 Aug 2007 #permalink

What about asking "What's the best way to learn?" and clarify whether we are talking about learning That (Paris is the capital of France), How (to ride a bike), Why (H2O becomes water), to Learn (become an effective learning agent)? Each of these categories are pretty different, and frameworks like Vygotsky's Zone of Proximal Development and the need for internalization seem to explain why spaced repetition makes sense...

It's "Harold" Pashler, aka Hal. Not Herbert. although I suppose some might think he's a total Herb....

By BikeMonkey (not verified) on 29 Aug 2007 #permalink

How does this apply to elementary aged students who take the summer off each year? Teachers often complain about their academic losses over the summer, and the need to cover old material at the beginning of the year to get them up to speed. Wouldn't the summer break be good for them, or is it too long? Should breaks only last a month, and be spread out over the year?

And why do you suppose that teachers have always told you not to cram? These findings are nothing new - learning psychologists produced 'real data' on effects decades ago, way back when the educational system was occasionally informed by science (following up on what teachers of language, music, swordfighting, basket-weaving, and tons of others must have known for hundreds). Any researcher touting this as new needs a reality check, and one less publication.

One thing which worked well for me (here I date myself) was to take my lecture notes, and on the same day, type them up on my manual typewriter. I found that this fixed things in my mind, allowed me to elaborate and add things which I had not written down, add references, and identify things which I did not understand. I found that I then needed little actual "studying for the test" when the time came. I have suggested this to students. It has helped some, but to my surprise, not all. There is also, I am told, literature that says that my practice is a waste of time.

One virtue of having learned something is that, even if you have forgotten all that you knew, you know that you can understand the matter, if you wish to make the effort.

By Jim Thomerson (not verified) on 30 Aug 2007 #permalink

I second Jim Thomerson - going over my lecture notes (literally writing them out again from scratch) at most a few days after the original lecture works very effectively. Since you've basically then studied everything twice, coming back to it before exams means you don't need to work hard at it again. Everything just comes flooding back.

I'd be curious to see some comparison of learning strategies for younger vs. older folks. Now that I'm middle-aged, I swear that the strategy of taking long breaks works better than it used to. And cramming, which worked well for me in college, doesn't seem as effective.

the same approach has always worked for me as far as written tests go...a variation on Jim Thomerson's method. I manually recopy my notes over in a more clear, organized fashion--and this seems to solidfy the information in my brain..and then the only studying I really have to is perhaps to re-read it before the test.
However, math problems were different for me..I have to re-work similar types of problems over and over...by the time the test rolls around it's almost like I have "memorized" the problem.

By laffingdukk (not verified) on 01 Sep 2007 #permalink

Another learning / training concept is, along with a spaced schedule of practice, to employ variable-priority training. This is where you incorporate several lessons together, and shift from one area to another. While it ultimately takes longer to gain expertise in a specific sub-topic, eventually the overall material is understood more deeply. This is something that's also been known for a while (see Danny Gopher and Art Kramer's work with Space Fortress), and seems to relate to the development of relationships between different aspects of a task. This is why it's good to learn how to drive in a car, where you're responsible for everything at once, than to spend a day focusing on the steering wheel, then a day focusing on the pedals, etc.