iStudy or iTest?

Last year, I (along with millions of other people) purchased an iPhone, and have been amazed at the diversity and number of available applications. Although much of the media focus has been on entertainment, lifestyle, and absolutely useless applications, my experiences at the Scheller Teacher Education Program at MIT have opened my eyes to the iPhone’s potential as an educational aid. While perusing the education section of the App Store for education programs, I found study aids for foreign languages, math, and geography, planners for homework and school projects, and several products that will likely increase the prevalence of cheating in the classroom. What caught my eye, though, were several digital flash card programs that were among the top downloads (e.g. gFlash+, Cram, iFlipr). These applications allow users to create their own custom flash cards from spreadsheets. Users can add images and text, have the cards appear in a random or sequential order, and share their cards with friends. An additional application feature found in these applications, and one that was regularly used by undergraduate students in the days of index cards (i.e. the stone ages), is that once the student correctly identifies the information on the flash card, that card is discarded from the pile. Thus, as testing progresses, the pile of flash cards is whittled down to a smaller and smaller set of unknown terms/concepts. I always regarded this as an efficient way to study: why test yourself on information you already know? In fact, I remember several instructors recommending that this study method be used. Although many use and espouse this method, a recent study by Karpicke and Roediger (2008) calls this approach into question. The two authors have long worked on the “testing effect”, the idea that the act of testing, in itself, influences learning and retention. In this study, they had undergraduate students study Swahili:English word pairs over a period of time. After studying, students were tested. This studying/testing cycle was repeated eight times over the learning phase, at which time nearly all of the students remembered every word pair. At this time, all students were asked to predict how many word pairs they would remember in a week’s time (on average, each student guessed 50%). Here is the interesting part: while the initial study session was identical, they set up four different study conditions for the rest of the periods: 1) Study all word pairs; Test all word pairs 2) Drop known word pairs from Studying; Test all word pairs 3) Study all word pairs; Drop known word pairs from Testing 4) Drop known word pairs from both Studying and Testing This design allowed the researchers to distinguish between the interactive effects of studying and testing with the review of known information. Traditional education models assume that repetition of encoding (studying) aids retention, while retrieval (testing) is neutral. In the learning phase of the study, students learned the same amount of information, in the same amount of time, across all treatments. One week later, however, student recall of word pairs was significantly higher when all word pairs were tested than when known word pairs were dropped from testing (over 80% recall with testing all words, vs. under 40% with dropping known words from tests). Amazingly, whether or not known word pairs were dropped from studying during the learning phase had no effect on information retention. It appears that repetition of the act of retrieval, rather than encoding, is what mattered in this experiment. While it may be argued that this effect does not hold for higher cognitive-level skills, the study design matches the way we expect undergraduate students to learn in many introductory biology courses. We give countless vocabulary words, cell parts, physiological steps, or species names to students, and ask them to recall that material on exams. I remember studying stacks of 300-400 index cards for a single exam in my first university biology class; I used the test-and-drop method, and prided myself on my efficiency at the time. It now appears that I was doing myself (and the friends whom I encouraged to use the same technique) a disservice. Perhaps I can still make amends: to all of you iPhone flash card users out there…make sure you turn that discard feature “off”.

Wondering how this applies to more complex concepts

This whole line of research showing that 'testing' is more important than straight studying is really interesting. In a lot of ways, it fits right into the theme that passive reading is not that effective a learning strategy. The interesting thing is that just trying to remember what you studied is enough to help at least memorization a lot. But the learning material here has a very simple structure - pairs of words. I wonder what the result would be for more complex material - for instance, biological concepts like natural selection or osmosis (to pick two examples from our company). What does 'testing' mean in that case - would simple recall testing like the ones in this study help people actually understand those concepts? Does active learning techniques like playing with simulations have the same effect? I'm guessing that just playing with directions with a simulation is not much better than studying. But if the student has to make predictions, or solve a problem themselves, it might be equivalent to the testing that they are talking about in this study. Would be a very interesting research question in its own right.

Post new comment

  • Allowed HTML tags: <em> <strong> <cite> <ul> <ol> <li> <dl> <dt> <dd>

More information about formatting options

CAPTCHA
This question is for testing whether you are a human visitor and to prevent automated spam submissions.
Image CAPTCHA
Copy the characters (respecting upper/lower case) from the image.

Find us at these events...