Why my test-related angst, you ask? Well, bear with me: I read a piece in the New York Times, which discussed some interesting-and potentially powerful-research on test taking as a way to learn. The intriguing aspect of the research is that test taking is not just for assessing learning-it can apparently promote learning, which is good news. The research, which originally was reported in Science, found that students who took a "retrieval practice" test recalled about 50% more of learned information after a week than students who used other learning methods, including just reading some text, studying a text in four consecutive brief sessions, or doing concept mapping, a much ballyhooed study technique involving drawing detailed diagrams linking ideas together

What is retrieval practice? Well, here is the way it was done in the study. First, the students read several paragraphs about a scientific topic (e.g., the digestive system). Then, without access to the text they read, the students wrote about what they could remember in a "freewriting" sort of essay for10 minutes. Afterwards, they re-read the text and then took a test on the material. After a week, the students took another short-answer test that measured (a) recall of facts and (b) ability to infer logical conclusions based on facts. Again, the students who engaged in retrieval practice did much better than those who used the other techniques.

Why does retrieval practice work so well? The psychologists cited in the Times piece believe that when we remember information, organize it, and get tested on it, we are creating meaningful connections among related ideas in memory. No surprise there. One surprising thing, however, is that the students who used retrieval practice were actually less confident about how they would do on the recall test a week later. Their earlier struggle learning the material clearly had decided benefits, but apparently it did not feel that way when retesting occurred. Yet, we know they out-performed the other study groups--maybe we should applaud this rare case of student modesty. I think we should admit that the results are not terribly dramatic--wouldn't anyone expect that writing about a reading (or anything for that matter) will lead to better retention of information?

Don't misunderstand me: I like these results--they are compelling, powerful, and do-able--and many teachers can introduce the technique into their classes. But is this approach practical? Moreover, is it desirable? How often will this form of test-retest learning be used? Will this launch a new and modified mania for "teaching to the test"? Perhaps students--and their teachers--should use such techniques selectively, for verifying the learning of key ideas in course material. That might be ok. Although I am in favor increasing the amount of writing that students at all levels produce, I would not want to increase the amount of tests they have to take.

Not so long ago, many psychology teachers--me included--shifted from lecturing to using other teaching and learning methods in the classroom. We tried and now use small group work, service learning, discussion, student presentation, in-class writing, journaling, critical thinking exercises, hands-on learning (in the case of psychology, having students collect and analyze data), and professionally oriented writing assignments (e.g., research papers, literature reviews). And some of us moved away from traditional tests--again, me included. In my case, I prefer to have students take in or out-of-class essay tests in lieu of multiple choice (recognition) tests in almost all my classes (and I admit my bias--I want my students to become better writers, so besides essay tests, they also write lots of short and some long papers in my classes). In other words, I give tests, but to me they are only a part of what happens in my classroom. Testing is by no means the most important thing I do.

I do worry that some people will regard retrieval practice as a great idea that should be introduced with dispatch into all kinds of classrooms. We need to remember that testing like this (and perhaps most testing) indicates what people recall; it does not necessarily reveal whether people know how to apply the knowledge to relevant situations, nor does it reveal creativity, problem solving, or any number of other characteristics. Psychologists like Robert Sternberg have been saying this for years. Tests have their place, but in the end a test is just a test--and only a test. We need to think about student learning in psychology, as well as our own learning, in much broader-and admittedly more interesting-terms.