This practice combines individualized (self-regulated) learning with collaborative problem-based learning to reap the benefits of both approaches in a single, completely online, graduate course in Educational Psychology.
The quizzing system used in this course for several semester has kept detailed records, allowing tracking of performance during a semester and across semesters. In general over 90% of students complete the chapters required for an A on that portion of the course. On average students pass about 70% of the quizzes they take, which seems to indicate that the quizzes are not too "easy." This percentage varies widely among individual students, from about 50% to 100%, indicating that the mastery and self-pacing aspects of the system are genuinely useful. Unfortunately, at this time there has been no opportunity to follow up to see how much of the material is retained over time. The PBL problems have also been successful, with some tweaking. It was found early on that students, even graduate students, do not always know how to collaborate, and this problem can be especially acute in an online setting. Detailed instructions for how to collaborate were developed, and these, along with feedback on the first problem, seems to have resulted in much higher quality problem solutions every semester. Students overall have indicated a high level of satisfaction with the courses, in spite of some frustration during each semester if they had trouble passing specific units.
learning effectiveness: The basic Learning Theories course in the College and Graduate School of Education, Health, and Human Services at Kent State was redesigned in 2004 as it moved to being an entirely online course. A central goal was to use instructional strategies from established theories of learning grounded on research showing their effectiveness. One challenge was to transfer these strategies to an online environment. Another goal was to use strategies that focus on different parts of the learning process, in this case on individual aspects versus group aspects. The two instructional strategies chosen were The Personalized System of Instruction (PSI), coming originally out of behaviorism (specifically operant conditioning) and Problem-Based Learning (PBL), derived from cognitive and constructivist learning theories. Both began several decades ago, and both have extensive research bases that appear to support their effectiveness in promoting learning and retention. At the same time, PSI focuses on individualizing instruction to the extent possible and PBL emphasizes group learning and the interplay of ideas among people. PSI was developed primarily by Fred Keller in the early 1970s. Among its key principles are self-pacing, mastery, activities as motivation, emphasis on writing, and the use of proctors. Self-pacing: Students should be able to proceed at their own pace through the material, although realistically there are likely to be some constraints on this. Mastery: Students have to master a unit before moving on. Most implementations of PSI include this element, although some disagreement may arise about the definition of mastery (100%, 90%, 80%). PSI is one of a variety of mastery learning systems that have been tried. Classroom activities as motivation: Instead of giving critical information during classes, PSI instructors generally allow the written materials to do that, reserving class time for motivational activities and other things. Emphasis on writing: PSI courses tend to have most critical information conveyed in writing. In addition, many of the communications channels between faculty and students are written as well. Use of proctors: Most PSI courses use graduate students or advanced undergraduates to proctor and score the tests. These proctors may tutor individuals at the same time. In a PSI course, students generally go through the course materials a unit at a time. If there is a textbook, each chapter is likely to be one unit. They study each unit, often with the aid of study guides, objectives, and other course elements. When they are ready (the self-paced aspect of the system), they take a quiz on the material for that unit. If they pass, they move on to the next one. If not, they study again and take another quiz until they do pass. The final grade, if one is given, is usually based on the number of units passed during the term. Students may have the opportunity to choose their grades by how they pace themselves through the material. Overall, research on PSI seems to show that students in PSI courses learn more and retain it longer compared to traditional lecture course. In contrast, Problem Based Learning (PBL) is explicitly a group-based instructional strategy designed to take advantage of the abilities of motivated and competent learners to discuss problems, develop and test hypotheses, and come to consensus while supporting the learning of everyone in the group. A PBL group might be given a problem by a facilitator who helps them define their needs for more information and problem-solving approaches. They may then do research individually on the background information needed and meet later to exchange that information and their ideas about the solution. Depending on the size and nature of the problem, the group may go through several cycles of research and group collaboration. Again, there is a research base that shows that PBL groups can learn as much or more, even on basic concepts (not just problem solving), than students taking traditional lecture courses. Both of these instructional strategies have recently been taken online; people are implementing PBL by using such communications tools as discussion forums, chat, whiteboards, and so on. To combine these approaches, we set up the following system for our learning theories course: Students had to use a well-known and high-quality textbook, and each of the eighteen chapters was a unit. To receive an A on the PSI portion of the course, students had to successfully complete 15 chapters, with many required and a few being optional. Successful completion was defined as scoring 90% or better on a ten-item quiz. Quizzes were taken online whenever students felt ready. They signed in and received a quiz with ten items generated randomly from a database of items for that chapter. The items included both multiple-choice items which were largely at the application level in Bloom's taxonomy and short-answer items. The original proponents of PSI advocated just the short answer items, but they are not machine-scorable, so the multiple-choice items were included as well. Because of the use of short-answer questions, the quizzes were not completely machine-scorable, so the student answers were both stored in a database and immediately emailed to the instructor. Feedback was not immediate but always took place within 24 hours. At two points while students were working on these units, they were assigned to a small PBL group. The first time when they completed the set of chapters on behaviorism; the second when they completed the set of chapters on cognitive psychology. At those points, the groups consisted of people who were at about the same point in the course (and hence had about the same background knowledge). The groups were given problems that encouraged them to apply and consolidate their knowledge of behaviorism and cognition respectively. Each group was given private online tools where they could work together, including a discussion board and a chatroom with a whiteboard. One of the problems often encountered with PSI is the failure of some students to pace themselves effectively. Deadlines were imposed for beginning and ending the PBL problems, which helped keep people on schedule through the semester. Throughout the semesters students were expected to participate in general online discussions about the course content, including individual theories and research findings, the application of these ideas to teaching, and other topics.
Most or all currently available course management systems have all the tools needed to implement these instructional strategies. A great deal of preparation must precede implementation to develop the item banks for the quizzes and create good problems for the PBL portion. During the semester, the instructor (or proctors, if available), must respond quickly to the quiz attempts.
Grant, L. K. & Spencer, R. E. (2003). The Personalized System of Instruction: Review and Applications to Distance Education. The International Review of Research in Open and Distance Learning. 4 (2). Retrieved online at http://www.irrodl.org/index.php/irrodl/article/view/152/233 April 26, 2006.