Here we are, at the end of another school year, and in schools all across the nation, teachers are giving comprehensive final exams and students are taking them. It’s a time-honored tradition, one that has been around for a hundred years or more. And even though you’d be hard pressed to find many students who say they actually enjoy taking finals, if you really pushed them on it, most students would say that finals are a necessary evil, something that has to be done in order to see what students have actually learned during the school year.

But do final exams really give us an accurate measure of what students have learned? And do they lead to new learning? The answer to the first question is a definite “Absolutely not,” and the answer to the second is, at best, “They might lead to a minor amount of new learning for a small minority of students.” When it comes right down to it, comprehensive final exams are nearly as much of a waste of time as standardized testing–and that’s saying something!

Cramming and Re-Cramming

OK, let’s backtrack a bit so I can explain where I’m coming from. Since the final exams at the end of the year are supposed to be “comprehensive,” that means that they’re supposed to cover all of the most important material covered across the school year (or across one half of the school year in schools where they give finals mid-way through the year and then again at the end of the year). This means that the material tested on the finals has already been tested at least once before, when the material was first taught.

So let’s look at those initial tests of the key material for a minute. What happened during that first go-around? Well, some students, I’m sure, actually studied the proper way the first time around (spaced out their studying, quizzed themselves instead of just re-reading their notes, asked the teacher for clarification of anything that was still murky, etc.). Students who went about their studying in this fashion more than likely learned the material well the first time around and scored well on the test. As a result, they likely created a solid representation of the material in long-term memory, which means that when the final exam comes along, all they have to do is access what they’ve already learned and put it down on the test. So, for students who truly learned the material the first time around, finals are a waste of time. We’re simply giving them two grades for learning the material once.

But now, let’s look at the much more common scenario. When most students took the initial test over the material, they probably did not study correctly. First of all, most students don’t know how to study correctly, and most teachers don’t know how to teach students to study correctly, but I’ve written about that extensively in the not-too-distant past, so let’s leave that aspect of it out of the equation for now and just focus on when students studied for the initial test.

You know the answer to this one, don’t you? Sure, because all of us have done it at one time or another–waited until the last possible minute to start cramming in the material in one massive studying binge. And we have over a hundred years of research that tells us exactly what happens when students cram for a test. If they cram hard enough, it’s possible that they were able to keep enough of the material revolving in working memory long enough to spit it out on the test and make a decent grade, and maybe even score top marks. This is why, when you ask students if cramming “works,” they almost unanimously say, “Yes!” It does work, if your definition of “works” is that it leads to an acceptable grade on the test.

But what happens in the hours and days after the initial test of the material? For those students who crammed, the material disappears rapidly. Hermann Ebbinghaus’s famous “forgetting curves” show that around 90% of the material that we don’t attempt to retain by further repetition will be gone within a week, and that virtually all of it will be gone in a month’s time. And do students go back and further study the material after the test on that material has been given? Not in my experience.

So now, here we are, a month or two after the first test of most of the material that will be on the final, and how much of the material do the “crammers” remember from the first go-around? Virtually none. And now that it’s time to test them on the material again, what do you suppose they’re going to do this time around? That’s right: cram again! Which means that (1) the scores these students receive on their finals are no more an accurate representation of their learning than they were the first time around, and (2) that these students will once again forget the vast majority of the material on the final within a week. So really, what’s the point?

Does this sound like a sound approach for promoting true learning and for getting an accurate measure of that learning? Hardly.

Creating a Culture of Retrieval While Taking the “Pop” Out of Pop Testing

So, what should we do instead? Well, when we look at the evidence from cognitive science, it’s possible to put together a set of guidelines for assessment that actually promotes student learning and gives us an accurate read on that learning without all the “false positives” we get as a result of students cramming.

Here’s how I would go about it. First, stop giving end-of-unit, end-of-semester, and end-of-year exams entirely. Just stop. When students know that a test sits out there at the end, they tend to procrastinate, saying to themselves, “The unit won’t be over for a couple of weeks yet. I can wait a while longer to start studying.” And, of course, more often than not, these students end up cramming on the final one or two days before the test, with the results described above. No, you’re better off not giving them the chance to cram at all.

How to do this? Simple. Give many, many more short, low-stakes, unannounced quizzes along the way, throughout the unit instead of the one big test at the end. I would give a 3-4 question quiz at the end of every class period over the key points covered that day, plus a question or two from previous learning–at the very least. On some days, I would also give short quizzes at the beginning of the period over the previous day’s material and perhaps also during the class period to check for understanding. None of these quizzes need to take more than five minutes.

Why is frequent quizzing helpful for learning? Because it forces students to retrieve information from memory. Research shows that having to recall information from memory strengthens the memory trace, making it much easier to recall the next time. This is called the “testing effect,” and it leads to much greater learning than do traditional review practices or studying by reading one’s notes.

One other key aspect of this approach is that I would never, ever tell students when a quiz was coming. Doing so simply leads us right back to the cramming and forgetting cycle.

How is this any different from the “bad old days” of “Gotcha!” pop quizzes? It’s different because you’re quizzing students all the time, once or twice per class period (2-3 times a period for block schedules, 6-8 times a day for elementary teachers in self-contained classrooms). In essence, you’re taking the questions you would normally give all at one whack on an hour-long traditional test, breaking them up into bite-sized pieces, and delivering them closer to the time of the initial instruction (from minutes after to a few days after).

And when students get used to this routine of constant checks on their learning, that takes the “pop” out of pop quizzing. They know they will be quizzed, and frequently. They just don’t know exactly when or over exactly what (though they know it will be soon and probably over material taught during that class period or in the recent past). This approach also leads to much better attentional focus, as students know that the quiz could happen at any time.

The Role of Feedback: A Second Opportunity to Learn

Now, let’s add an element of feedback to the equation to further super-charge the learning.

In the traditional test-at-the-end-of-the-unit approach, there’s often very little feedback when the tests are given back beyond some red check marks and a score, though some teachers may put in the extra effort to provide comments in the margins. But even if they do provide well-intended extra comments, how often do students take the feedback given and use it to improve their grasp of the content or skills covered on the test? Almost never, of course. I say, “Of course” because this system, as set up, doesn’t reward students for looking at and acting on the feedback. The test is over. The score is in the grade book. Will going back to review what was missed and take another crack at learning it improve one’s grade? Nope. So why do it?

In the system I’m proposing, however, we can add a feedback loop quite easily and give students more than one chance to master the material. Here’s how I would go about it:

First of all, ask questions about key material more than once, on several different quizzes. So, at the end of today’s class period, I might have a question on the quiz about a key piece of information. A couple of days later, I may once again ask a question about this key point, mixing it in with questions about that day’s content. Then, I may bring that question back once again a week or so after initially covering that material in class. Not only does this approach give students multiple chances to learn key content, but it also forces them to retrieve key material multiple times, leading to strong memory traces and true long-term memory of the content.

In addition to asking questions multiple times about key content, I would build in a feedback loop to the process. For example, when students take a quiz, I might provide a simple “exam wrapper” for the quiz (check out this link for information about exam wrappers). In this case, it might be a simple, “If you missed question #1, look back at pages 235-236 from your textbook, as well as your notes about __________. If you missed question #2…, etc.” This approach is preferable to going over the quiz items in class, for at least two reasons: (1) the exam wrapper approach off-loads the feedback to a homework assignment, so no extra class time is taken up by going over the quiz; and (2) having students look back at the proper material individually focuses them on just the items they missed, whereas going over the quiz in class wastes time (if a student missed one out of four items, for example, the review of the three items he didn’t miss is wasted time for that student).

Of course, if you notice a trend in the quiz results–for example, a high percentage of students miss the same item–it would be a good idea to go back over that material. So, the feedback loop works for both teachers and students (as opposed to the feedback a teacher gets in the traditional system–too late to do anything about missed learning).

Three Ways to End Your Units/Semesters/School Years

Some of you may be wondering at this point, “If I follow the approach described here, what will I do at the end of my units of instruction to replace the traditional test?” or “What will I do at the end of the school year to replace the traditional final exam?” So, I’ll end by giving you a few ideas for that.

1. Presentations of Long-Term Projects: If you follow the approach I’ve laid out here, you’ll be evaluating your students’ learning as you go, and you’ll be getting rid of those end-of-unit and end-of-year tests. One great way to end a unit or the school year is to have students work all along on long-term individual or cooperative projects and finish off the unit or school year with the presentations of these projects. There are two main reasons that long-term projects are much better to end a unit or year with than a test: (1) since students work on them a little at a time over an extended period, there’s no cramming involved; and (2) by working with material over an extended time, every time students return to their projects to work on them, they’re reviewing the material they’ve already found, doing more research, reorganizing their material, etc. These multiple exposures to the material virtually guarantee that students will commit the material they work with in their long-term projects to long-term memory.

2. Reflection: Another great way to end a unit or school year is to have students look back over the period of time in question and reflect on what they’ve learned about the content and about themselves as learners. If you use some type of portfolio system to capture samples of student work over time, they can look over their work and make judgments about quality and set goals for improvement.

3. Celebration: There seems to be a woeful lack of celebration in schools, and I don’t understand how that can be the case. After all, every day our students are learning things about the world that they didn’t know before or mastering skills they didn’t possess before. I can’t think of anything more worthy of celebration, can you? The end of a unit or school year is a great time to simply cut loose a bit and say, “Hey, you guys did that. Let’s party!” (in moderation and within school guidelines, of course).

Make a regular habit of ending each unit throughout the school year (and the school year itself) with one, two, or all three of these, and you’ll find that these are much more satisfying ways to “put a bow” on things than ending with a test.


As you can see, it isn’t testing itself that’s a waste of time, but the way we’ve gone about testing in the past. We need to re-frame the idea of testing, not as a “gotcha” situation with little or no timely feedback, but as an opportunity to frequently check in on students and reinforce what they’re learning. The approach I’ve laid out here would virtually eliminate cramming and massively boost retention of material. Who wouldn’t want that?

I hope I’ve given you some food for thought today. I know it’s probably too late to put any of this into practice this school year, but think about it over the summer, and if you’re ready to turbo-charge your students’ learning and quit wasting time on out-moded testing practices, you can start next school year off right from the beginning!