W4 Practice Testing - PDF

Slide 1: Welcome to week four. During week four, you are going to learn about some effective study strategies and then also about some popular but likely not so effective study strategies. As the title here already hints to, reading the text multiple times may not be the most effective study strateg...

Slide 1: Welcome to week four. During week four, you are going to learn about some effective study strategies and then also about some popular but likely not so effective study strategies. As the title here already hints to, reading the text multiple times may not be the most effective study strategy, even though it is a very popular one. Slide 2: What does learning science tell us about effective study strategies? Here are five that are proven to be effective. At the bottom of the slide, you see a reference to Dunlosky, Rawson, Marsh, Nathan and Willingham (2013) paper titled "Improving students' learning with effective learning techniques: Promising directions from cognitive and educational psychology." This paper is available in Leganto in multiple folders and most of the evidence I present in Week 4 is from that paper. Dunlosky et al. say that practice testing and distributed practice are of the highest utility, because they benefit learners of different ages and abilities, and they have been shown to improve students' performance across different kinds of tasks and educational context. Interleaved practice, elaborative interrogation, and self-explanation were assessed as having moderate utility. The reason for a lower utility rating was that the benefits of these techniques have been shown to generalize across a more limited set of tasks and age groups, in other words the evidence for the efficacy is more limited. We will examine these limitations in more detail in each of the videos. Slide 3: In this video, we will focus on practice testing, also known as retrieval practice. Practice testing is likely not your, or your future students, favourite learning activity, but it is one of the most effective learning activities you can do. Note that when we talk about practice testing, we mean testing that is low stakes or no stakes practice or learning activity. It is testing that is not marked or is marked as pass-fail, but doesn't count towards unit’s final mark. For example, the quizzes you do in iLearn are an example of practice testing that are marked pass or fail. They also control that you have achieved sufficient level of knowledge or mastery of the content to move forward. Your assignments, in contrast, do not count as practice tests because the stakes are higher as they count towards your final mark. Practice testing works because learning is about understanding, generating a memory trace and being able to access the memory trace. You need to practice understanding, remembering and accessing the memories. Typically, we practice understanding and remembering, or making the memory traces, but not accessing them. Slide 4: Practice testing is about making the memory traces more accessible. It is a learning activity, not an exam, that you can complete on your own or with peers. Practice testing can, for example, involve practicing recall of target information via the use of actual or virtual flashcards, completing practice problems or questions included at the end of many textbook chapters, or completing the quizzes included in iLearn. It can involve you reading a page, then looking away from the page, and trying to recall as much as you can of what you just learned. If you're studying with a peer, practice testing could involve taking turns in asking questions about what you just read. When there is a criterion for passing, as in this week's learning activities, then it becomes a mastery learning activity. Mastery learning is an old idea initially proposed by Benjamin Bloom in the late sixties (you may have heard of Blooms taxonomy, and most certainly will hear about it). In schools’ using mastery learning, the typical model is that students must achieve some predetermined level of mastery before they move on to new material. This is particularly important with content like maths where new learning builds on old learning and gaps in knowledge base can prevent learning the new material. Typically, if students fail to achieve mastery, they are then provided additional support, tested again, and so on until they pass and move on. Slide 5: Here are some suggestions from your textbook authors on how to do retrieval practice. At the end of this video I will tell you how to access this information, and whole lot more on their website. Put away your class materials and write or sketch everything you know. This is called free recall. Your sketch could look like a concept map similar to the concept map we have been using to define learning, or it could look like something else – I used to study with a then young man who drew cartoons as his notes. Whatever you do, you need to be as thorough as possible. After your response is ready, only then you check it for accuracy and important points you missed. Finally, once you know what you missed you can then review only those points, but not the whole text as that would be wasting your time. Slide 6: Second, take as many practice tests as you can get your hands on. If you don't have ready-made tests, try making some of your own and trading with a friend who has done the same. This is a great way for you to help each other to learn the material better. Slide 7: And third, you can also use flash cards. If you do, make sure you practice recalling the information on them and go beyond definitions by thinking of links between ideas. The one possible problem with flash cards is that they segment the knowledge into independent bites and you may not build the connections between the definitions in different flash cards. As you can see here in the last picture, the gentleman is organizing his flash cards into some kind of a hierarchy, similar to what you saw, for example, when you were learning about long-term memory. Slide 8: We know that practice testing works. More than 100 hundred years of research has yielded several hundred experiments showing that practice testing enhances learning and retention. Already in 1906, Edward Thorndike wrote that, "The active recall of a fact from within is, as a rule, better than its impression from without." With this, he could have referred to the difference between practice testing and re-reading. You will hear about Thorndike a little bit more later on in this unit. Slide 9 (Fig 7 and Fig 8): Let's look at two example studies. The results on the left are from a study by Spitzer completed in 1939, that is 80 years ago. They suggest that practice testing effect is not limited to better performing students. Spitzer first gave 3,600 year six students a baseline reading comprehension measure. After that, he asked all students to read a new text. Half of the students completed a practice test immediately after reading the new text, and the other half did not. After that, all students completed a multiple-choice test, half of them one day and the other half seven days later. Spitzer reported final test performance separately for the top and bottom third of performers on the baseline reading comprehension measure. As you can see in the figure, taking the practice test benefited both students who did well on the baseline measure, as well as students who did less well on it, although it could be that the benefit was larger for the better performing students with about 20% difference versus 12% difference between the groups. Slide 10: The results on the right are from a much more recent study by Butler. The title of the study is telling: Repeated testing produces superior transfer of learning relative to repeated studying. Butler was mainly interested in transfer of learning to answering different kinds of, but related questions, in experiment 1B here. There was also an Experiment 1A that assessed simply retention of material after practice test and restudy, and it showed large differences in favour of practice testing as well. In experiment 1B, Butler wanted to test whether the positive effects of practice testing are true for both factual information and conceptual information. He presented undergraduate students with expository texts for initial study, which was then followed either by restudy of the texts, or by a practice test, tapping the key facts and concepts from the texts. One week later, performance on new inference based short answer questions focusing on the key facts and concepts was assessed again. You can see here on the first columns that performance was much better for students who did practice testing than for students who studied the material twice. Experiment two essentially repeated the results from experiment 1B, but now students in the restudy condition studied repeatedly isolated facts and concepts, as in having different pieces of knowledge in flash cards. Butler made sure that students in both conditions spent the same amount of time either studying or answering the practice questions. As you can see in middle columns, the differences were equally striking in favour of practice testing. The outcome of experiment 3 is particularly interesting. Here, the criterion test involved far transfer in that questions required the concepts from one domain to be applied in another domain. Students had to apply information they learned about bat wings to make inferences about the development of new kinds of wings for aircraft. Again, the difference is striking. On the basis of these experiments, Butler concluded that the findings clearly demonstrate the effectiveness of retrieval practice in promoting both retention and transfer of knowledge. There are literally hundreds of studies showing the same kind of an effect with different age groups and learning tasks. Even if we control the time spent studying, practice tests work better than restudy. If we don't control the time spent studying, typically students in restudy conditions spend much more time than students in practice test conditions, but gain less knowledge. There is a reason to use practice tests. Slide 11 (When do practice tests work?): The question here is: Under what conditions do practice tests work? The answer is that pretty much all the time. It doesn't seem to matter what kind of practice tests we use. Cued recall, including multiple choice quizzes, free recall, short answer questions, fill-in-the-blank questions: they all result in better performance. It is possible, however, that free recall practice may help the most. It also seems that all dosages work, but more tends to be better, particularly if the practice tests are spaced over time. You will learn about spacing effect later this week. All students studied so far seem to benefit from practice testing and studies have been completed with preschoolers, primary school students, high school students, university students, and older adults. Practice testing seems to work whether we are learning from texts or videos, or whether we're learning spatial information like maps, for example. They also work whether we need to learn factual information or whether we need to comprehend something more complex or transfer the knowledge into new domains and to real world problems. In sum, the positive effects of practice testing have been shown with a wide variety of participants and learning tasks. Slide 12: Why do practice tests work? Well, first you practice recall. And when you do that, you create connections in your memory between the information you try to recall and other related information. When you try to recall the information again, in the actual testing situation, you now have more access routes to that information. When you practice recall of specific information, you also tend to organize the information that comes to your mind during that process more effectively. You build new connections and you do this with no conscious effort. In sum, practice testing enhances both memory and access to memories. Your memories are better organized and they are easier to recall. Slide 13 (Summary): Here is a summary from Dunlosky et al. "On the basis of the evidence described above, we rate practice testing as having high utility. Testing effects have been demonstrated across an impressive range of practice test formats, kinds of material, learner ages, outcome measures, and retention intervals. Thus, practice testing has broad applicability. Practice testing is not particularly time intensive relative to other techniques, and it can be implemented with minimal training. Finally, several studies have provided evidence for the efficacy of practice testing in representative educational contexts." Slide 14: And here are a few more tips before we finish. Retrieval practice works best when you go back and check your responses for accuracy afterwards. This way, you review the answers to the questions, not all of the material. If you are struggling, identify the specific topics, concepts or facts you are struggling to remember and work on those alone with the materials closed. Don't recall only words or definitions. Make sure you recall main ideas, how those ideas are related or different from one another, and think of new examples, how you can apply those ideas. Slide 15: Here are some additional readings. The authors of your textbook have a website called Learning Scientists. On that website, they have a blog, and on the blog, they have all kinds of useful short explanations of these same concepts, including several explanations written by teachers on how to use these learning strategies in a classroom environment. With practice testing that is easy, but with some of the other strategies it may be less obvious. Here is one link for how to practice retrieval, a second for the concept maps, similar to what you have on page 121 in your textbook. The last one explains how to use flash cards effectively. You also have research summaries available in Leganto. You have a very accessible summary from Dunlosky written for teachers. You have a little less accessible summary from Dunlosky et al. written at the same time. And then you have a quite accessible review from Weinstein, Madan and Sumeracki from 2018, written at the same time when Weinstein and Sumeracki were writing the book. There are also a few science translation articles in Leganto. I think all of these are from Scientific American: The Science of Learning, Forget Cramming, and Why Testing Boosts Learning. Those are all quite readable and easy access papers. You should start with Dunlosky, 2013, and with the translation articles, before you go into the Dunlosky et al. paper, or Weinstein, Madan and Sumeracki paper. The blogs are very accessible as well. And useful. That is all for the first video of week four. In the second video, we'll examine distributed practice, but before you can access that video, you need to complete the quiz for this video. Thank you.