Designing Effective Multimedia for Physics Education

multimedia on nuclear reactor physics download and also multimedia in physics teaching and learning
MarthaKelly Profile Pic
Published Date:12-07-2017
Your Website URL(Optional)
Designing Effective Multimedia for Physics Education A thesis submitted in fulfillment of the requirements for the degree of Doctor of Philosophy by Derek Alexander Muller School of Physics University of Sydney Australia 2008Abstract This thesis summarizes a series of investigations into how multimedia can be designed to promote the learning of physics. The ‘design experiment’ methodology was adopted for the study, incorporating different methods of data collection and iterated cycles of design, evaluation, and redesign. Recently much research has been conducted on learning with multimedia, usu- ally from a cognitive science perspective. Principles of design developed in this way have not often been tested in naturalistic settings, however. Therefore in one preliminary investigation students’ perceptions of a popular science video were investigated. Opinions aligned well with most principles though areas for further research were identified. In order to understand the challenges and opportunities presented by physics teaching, a survey of all lecture courses on the topic of quantum mechanics was undertaken. The lectures were a sophisticated form of multimedia, however inter- activity in all lectures was low. The learning that results from this teaching was evaluated using a questionnaire on quantum tunneling, a key quantum mechanical phenomenon. The survey re- vealed that students had many alternative conceptions on the topic and that these could be grouped into a small number of alternative answers. This finding is similar to many of the findings from science education over the past three decades. Using this background, two multimedia treatments were developed to teach the topic of quantum tunneling. One consisted of a lecture-style explanation with only correct information presented. The other took the form of a dialogue between a tutor and student, involving several of the common alternative conceptions. Students who saw the Dialogue performed significantly better on the post-test than those who saw viiithe Exposition. In order to generalize the findings, four multimedia treatments on Newton’s first and second laws were created and evaluated in a similar way. A refutationary treat- ment, in which alternative conceptions were stated and refuted by a single speaker, and an Extended Exposition treatment were evaluated in addition to the Dialogue and Exposition. The Dialogue and Refutation outperformed the two expository treatments, confirming the benefits of including alternative conceptions. In a third iteration of the design experiment, four Newtonian mechanics treat- ments were evaluated with a new cohort of students. The Extended Exposition was replaced by a Worked Examples treatment in which important details were repeated to solve numerical problems. Cognitive load was directly measured in this exper- iment. Results showed that treatments containing alternative conceptions involved higher cognitive load and resulted in higher post-test scores than the other treat- ments. ixChapter 1 Framing the study During the course of this doctoral thesis I was awarded a grant to develop multi- media for senior high school physics students. In the initial stages of the project, my team of teachers, academics, educational technologists, and I selected syllabus topics that we thought would benefit from additional multimedia resources. We planned to develop tools that could be distributed freely over the Internet for stu- dents to use in their own time or for teachers to show and discuss in class. Searching the physics education literature we honed in on topics that students often find con- fusing. We noted common misconceptions and the methods that have shown some success in achieving conceptual change in the classroom. Through web searches and discussions with educators we determined which physics concepts were thor- oughly covered in textbooks and online and which had few or unclear resources. The only question that remained at the conclusion of our preliminary analysis was: once the learning objectives are specified, how does one produce effective multimedia for physics education? This question had been the focus of my PhD research and now I was faced with it in a very real sense. The problem of designing multimedia to promote learning is a common one yet it has received uneven and, until recently, inadequate attention from academics. This thesis represents one at- tempt to understand the challenges and opportunities presented by multimedia for the learning of physics. Although the term multimedia has many definitions and connotations, in this 1thesis, it refers to any presentation that combines words and pictures to form a co- herent message. Illustrated books, lectures that include diagrams, and animations with narration all therefore constitute multimedia. This might seem like an incred- ibly broad definition, but there are good theoretical and empirical reasons for col- lecting such a diverse range of presentations under one banner (Clark 1994a, Mayer 1997), as I will explain in this chapter. 1.1 The rise of multimedia Since the development of language only a few hundred thousand years ago, multi- media has grown in its sophistication and availability. The first recorded multimedia probably consisted of hieroglyphs and paintings on stone tablets. It was undoubt- edly time-consuming to create, and comprehensible only to scribes and scholars. In the centuries that followed, multimedia was rare and accessible only to the educated upper classes. Several inventions led to significant advancements in the development of mul- timedia. One was the printing press, which, after 1450, allowed large volumes of text to be readily copied and distributed. Lithography, a similar technique for print- ing images, was developed in 1796. Photography followed in the early nineteenth century. These inventions made text and accompanying pictures available to more of the population at less expense. The development of the motion picture marked another important milestone at the end of the nineteenth century. The quick succession of static images created the illusion of motion with explanatory text interspersed at intervals throughout the movie. The early twentieth century saw the invention of ‘talking pictures,’ with speech and sounds synchronized to the action in the film. This meant that literacy was no longer a barrier to understanding multimedia. Television and video permitted a similar experience to film, but at less expense and with greater flexibility. Again, words and pictures became more widely avail- able. Digital video discs (DVDs) and interactive video discs provided higher quality sound and images, and added an element of interaction between user and multime- 2dia. Finally, over the past two decades, the computer has been transformed from a calculator and word processor into a multi-faceted multimedia communication device. It goes without saying that we are currently experiencing the greatest information- sharing explosion in human history. Multimedia availability is increasing at an ex- traordinary rate with new online repositories and distributors appearing daily. The Internet is maturing, bandwidths are expanding, new software is being created, and hard drive capacities are on the rise. It has never been easier to create or distribute multimedia. Video, one of the most common multimedia formats, has become so widely available that a popular website regularly handles over 100 million video downloads per day (BBC News 2006). The objective of this thesis was to investigate how multimedia can be designed and used to promote the learning of physics. I focused mainly on the video form because it encompasses most attributes of other multimedia, however, this research should have implications for other approaches. Because multimedia is being used at all levels of education, studying and im- proving its effectiveness is a significant and worthwhile challenge. It would be ideal if students could learn about science by working in groups, devising and performing experiments, and discussing their ideas with knowledgeable, experienced teachers. However, resources are limited and students must often learn by themselves with textbooks, videos, and online multimedia. Furthermore, after leaving formal ed- ucation, learners must be able to build on their knowledge with different types of learning resources. There are additional reasons for studying multimedia. Almost all learning ex- periences, whether interactive or not, consist of segments of linear multimedia; ex- amining this building block can arguably provide insights into more complicated pedagogical methods. In addition, multimedia provides a confined arena in which to test different instructional strategies with large cohorts of students in real learning environments. Novel teaching implementations in science education have been crit- icised for varying several aspects of instruction simultaneously without attempting to understand the features essential to their success (e.g. Guzzetti, Snyder, Glass & 3Gamas 1993). Multimedia offers a transparent and repeatable way to study specific aspects of the teaching and learning process. 1.2 The research that wasn’t there Given that people have been using multimedia in education for decades, it seems reasonable to expect a sizable body of research to exist on how it may best be designed. Unfortunately, this does not seem to be the case. At the outset of this re- search, Moore, Burton & Myers (2004) summarized in their review of the topic that “with few exceptions there is NOT a body of research on the design, use and value of multimedia systems” (p.997, emphasis in original). Although the exceptions re- ferred to in the preceding quote form the theoretical foundations of this thesis (see Chapter 3), it is startling that a century of research and use of educational technol- ogy has yielded so few productive outcomes. This lack of research is readily apparent in the literature. A study in the Ameri- can Journal of Physics illustrates the types of questions that have been asked repeat- edly in educational technology studies, with little success. Lewis (1995) explored the impact on students’ grades and attitudes of replacing standard tutor introduc- tions to experimental laboratories with video introductions. The videos did not include anything that was not part of the usual tutor presentations. Perhaps unsur- prisingly, the researcher found that students’ marks were the same with the videos as they were with tutor introductions. Accounting for this result, Lewis specu- lates, “it may be that the video medium is unsuitable for the purpose of laboratory introductions or that the particular videos used here were deficient in content or presentation” (p.469). He further supposes that it may be the ‘passive’ nature of video that limited its effectiveness. He does not consider, however, the possibility that one standard multimedia presentation may be as good as another. Why should one expect a video to outperform a tutor, presuming the tutors are knowledgeable and readily available during laboratory? In another study, Rieber, Tzeng & Tribble (2004) measured the learning about Newtonian mechanics that resulted from several different instructional treatments. 4All students interacted with a computer simulation in which the goal was to move a frictionless ‘ball’ to a specified target. Half of the students received graphical feedback while the other half received textual feedback. In addition, only half of the sample received brief multimedia explanations of the physics involved, inter- spersed throughout the simulation. The best-perfoming group on the post-test was the graphical feedback with multimedia explanation group. Without a multimedia- only control, the authors conceded “the issue of how much learning is taking place just by having participants view the explanations without participating in the simu- lation is open to question” (p.321, emphasis in original). Kim, Yoon, Whang, Tversky & Morrison (2007) investigated student learning about bicycle pumps using multimedia materials. Still graphics were presented un- der four conditions: (1) all at once, (2) successively, (3) self-paced, or (4) animated. It was thought that the animated materials might have a superior effect because they could be seen as “more interesting, aesthetically appealing, and therefore more mo- tivating” (p.261). Presentation mode did affect student perceptions of the materials, including interestingness, enjoyment, and motivation; however, comprehension test scores did not differ among the groups. The three studies outlined above are symptoms of a body of research that has failed to establish answers to fundamental questions about learning with multime- dia. Aspects of these studies typify the difficulties with educational technology re- search and help understand why a more relevant theoretical base hasn’t been estab- lished. During the twentieth century, film, radio, television, video, and computers were all introduced into classrooms at different times. The patterns of their imple- mentation, use, and supporting research bear striking similarities, with all technolo- gies failing to live up to expectations. Research on these educational technologies did not establish a general and ro- bust theoretical foundation for designing multimedia for several reasons: 1. The advantages of new technologies were seemingly self-evident. So much hype accompanied each innovation that rigorous research was seen as unnec- essary. 2. The questions asked by researchers were generally media-comparative and 5disconnected from theoretical considerations. 3. The practical drive to introduce technology into schools limited the time in which research was carried out, transferring the burden of using and proving the efficacy of new inventions to designers. 4. Fundamental perspectives on how people learn shifted continually over the past century. These four points are addressed in the sections below. 1.2.1 Technology: The obvious solution to our problems The implementation of any new technology into education has typically begun with incredible rhetoric and expectations. Marketers and technology developers have fo- cused on the ground-breaking abilities of the new technology to promote interest in its application to the educational domain. Thomas Edison’s appraisal of the mo- tion picture is an oft-cited example of the excitement that accompanies innovation. Promoting his invention, he proclaims “that the motion picture is destined to rev- olutionize our educational system and that in a few years it will supplant largely, if not entirely, the use of textbooks,” (Edison 1922 as cited in Cuban 1986, p.9). Such claims are not restricted to a bygone era, however; for example Semrau & Boyer (1994, p.2) note “the use of videodiscs in classroom instruction is increas- ing every year and promises to revolutionize what will happen in the classroom of tomorrow.” Clark & Estes (1999) attribute the ineffectiveness of past research pro- grams, at least in part, “to a history of mindless and demonstrably wrong advocacy of popular electronic media to foster motivation and learning” (p.5). Another common element of marketers’ campaigning is a contrast between the promises of new technology and the existing state of education. Pessimistic claims about the school system have been routinely juxtaposed with the dramatic prophe- sies for future technologies. For example, Edison took aim at textbooks. I should say that on the average we get about two percent efficiency out of schoolbooks as they are written today. The education of the future, as I see it, will be conducted through the medium of the motion picture 6...where it should be possible to obtain one hundred percent efficiency. (Edison 1922 as cited in Cuban 1986, p.9) Where Edison comes up with the figure of two percent for the efficiency of text- books is unclear, as is the notion of one hundred percent efficiency, but his argu- ment was understandably a persuasive one for politicians and the public alike. In his time, the technological revolution was in full swing and people were eager to consider the concept of efficiency as it related to agriculture, steam engines, and education. The excitement surrounding new technologies diverted attention away from rig- orous research. Researchers and the general public were intuitively convinced of the effectiveness of new inventions. “Their reasoning seems to suggest that if re- search does not find evidence for something that seems so powerful, then research as an inquiry strategy must be flawed” (Clark & Estes 1998, p.5). 1.2.2 Is this medium better than the other one? Researchers adopted the perspective that educational efficiency could be measured and optimized, and began to investigate the intrinsic advantages of one medium over another (Russell 1985, p.47). The medium itself seemed the obvious variable for investigation, rather than the experience of the learner. McLuhan’s (1964) refrain ‘the medium is the message,’ focused attention on new and exciting inventions, fu- elling the technology-centred approach. Early studies compared the performance of students who watched an instructional film to those who received only traditional lecture instruction, in experiments similar to Lewis’s (1995) study. The results showed increased motivation among students who watched films and either supe- rior or equivalent academic performance compared to a control group (Cuban 1986). Excitement due to novelty, methodological confounds, or a Hawthorne effect likely account for much of the success of these studies (Clark 1983). Similar research on educational television showed impressive results, increasing math, science and reading scores on standardized tests. However, researchers did not ensure the so- cioeconomic statuses of different treatment groups were comparable and failed to 7report any differences in this measure (Cuban 1986, p.35). Even when comparative research showed negative or no significant difference results for new technology, its promoters used the enthusiasm, assumptions, and excitement surrounding the technology as effective counter-arguments. When Clark (1983) concluded that no particular media had a unique impact on learning and that research seeking such an impact should be abandoned, he be- lieved the point to be uncontentious and well-supported by the evidence. “Media are mere vehicles that deliver instruction but do not influence student achievement any more than the truck that delivers our groceries causes changes in our nutrition,” he wrote (p.445). The paper kicked off debate in the research community because, stated explicitly or not, the notion that media inherently affects learning had been a presupposition of virtually all previous studies in educational technology. Many researchers debated the claim, although perhaps the best articulation of opposing viewpoints is contained in the writings of Clark and Kozma (Clark 1988, 1994a, 1994b, Clark & Salomon 1986, Kozma 1991, 1994a, 1994b, 2000, Kozma & An- derson 2002). Kozma (1994b) argued that different media have particular capabilities, which enable different learning experiences. “A particular medium can be described in terms of its capability to present certain representations and perform certain opera- tions in interaction with learners who are similarly engaged in internally structuring representations and operating on these” (p.11). Clark (1994a) maintained that the learning experiences in any form of multi- media could be made almost identical to any other with adequate preparation. For example, the technique of zooming used in video to focus on a component of a larger system could be illustrated diagrammatically with a magnification bubble. Based on this interchangeability, he proposed the ‘replaceability’ challenge: “to find evidence, in a well designed study, of any instance of a medium or media at- tributes that are not replaceable by a different set of media and attributes to achieve similar learning results for any given student and learning task.” The challenge was meant to demonstrate the equivalence of different platforms and highlight the methodological differences that actually impact on learning. 8It cannot be overlooked that the history of educational technology, with new technologies repeatedly delivering much less than they had promised, seems to bear out Clark’s argument. After a technology’s initial implementation into schools, interest has waned and its use has declined. For example, instructional televi- sion and computers occupied at most four to eight percent of instructional time, even in well financed schools with professed interests in implementing technology (Cuban 1986). Thus, it is understandable when critics claim technology has failed to live up to expectations (Tyack & Cuban 1995). If there were even one instance where a particular media afforded a unique and profound benefit over competing technologies, would it not be widely adopted and repeatedly cited as evidence by media proponents? 1.2.3 Implementing technology in schools Although the outcomes of early media research may have been dubious, they were sufficient to encourage educational administrators to implement new technologies in schools. This worsened the research deficit as technologies gained the appearance of maturity and academics became experts in a field with little theoretical or empirical basis. No sooner had my colleagues and I begun exploring the potential use of the computer for teaching science than colleges began offering Master’s degrees in computer education. Although no one had any knowledge or experience using computers to teach anything, experts were instantly trained, hired, and funded to bring computers into the public schools. (Cromer 1997, p.108) Similar observations are made to this day of university educational technology pro- grams that de-emphasize scientific research and work under the assumption that technology is inherently beneficial (Clark & Estes 1998). Most recently school districts have invested incredible sums of money to bring computers into the classroom. 9The adoption of microcomputers by U.S. schools has been explosive, going from essentially zero in 1980 to better than one for every nine- teen students by the early 1990’s. The computer’s educational roles change from year to year, as their functionality evolves, and today their purposes are as unclear as they are unquestioned. (Cromer 1997, p.121) The rapid introduction of technology to education has been witnessed many times. In the 1930’s, with the drop in price of radio receivers, governments invested heavily in educational radio broadcasting. A majority of schools bought into “the textbook of the air” and purchased at least one receiver set (Cuban 1986, p.19). With a particular technology available in classrooms almost from the date of its invention, instructional designers have carried the burden of the outlandish promises made by technology salesmen. To this day, designers must make do with what little reliable literature is published and use intuition or industry rules of thumb to make the balance of decisions. These designers are also the only qualified sources to write textbooks on instructional design based on their experience. Blinn (1989) outlined a number of design criteria for educational animations, drawing on his experience as an animator for a physics education video series. Although likely very useful, these guidelines are a starting point for investigation rather than established princi- ples of best practice. Yet these types of resources have been the only references for designers making costly decisions about how to create multimedia. Clark & Estes (Clark & Estes 1998, 1999, Estes & Clark 1998) have classified educational tech- nology produced in this way as ‘craft’ solutions, uninformed by scientific research. They claim that it is the lack of concrete theoretical foundations and the subsequent proliferation of craft technologies that has led to the unreliability of technological solutions. These craft solutions are the most common type of educational technology, and, since they are not developed or evaluated scientifically, are unable to directly inform the body of research on learning with technology. This perpetuates the cycle of craft educational technology, further inhibiting progress in the field. 101.2.4 How do learner’s learn? Another factor that has limited the development of multimedia research is the shift- ing perspectives in the educational literature of what constitutes learning and how it is achieved. Early behaviorist research rejected the notion of cognitive entities and focused instead on observable actions. Under this paradigm, punishment and rewards were used to modify behavior. In the 1960’s this view was displaced by cognitivism. However, within this movement, different branches of research have formed with little overlap between groups. With social and radical constructivism, information processing models, connectionism and associationism, it has been in- credibly difficult to form a coherent body of knowledge. In media research, very few studies were based on theoretical frameworks that accounted for the effects of technological interventions (Clark 1994a). When learn- ing theories were employed, they were dependent on a delivery model of education (Kozma 1994b, Clark & Estes 1999). Media were viewed as delivery vehicles, per- mitting the question ‘does this medium deliver information more efficiently than other media?’ In fact, it has only been in the last decade that researchers in the field have moved to a constructivist paradigm. “It is time to shift the focus of our research from media as conveyors of methods to media and methods as fa- cilitators of knowledge-construction and meaning-making on the part of learners” (Kozma 1994a, p.13). 1.3 The equivalence principle In sum, until recently volumes of educational technology research have yielded lit- tle theoretical or practical guidance for the design of multimedia. It is arguable that the most significant conclusion yielded by previous studies is that the learn- ing experiences with and without a particular technology can be made equivalent with adequate forethought. On this matter, it is worth pursuing an analogy with an equivalence principle from physics that is central to the theory of General Relativity. When it dawned on him, Einstein called this principle his happiest thought. 11For all of human history before the twentieth century, gravitational forces were perceived quite separately from the concept of acceleration. A person experiences a strong gravitational force any time he or she is in close proximity to a large mass, as is the case on the surface of the earth. Acceleration, on the other hand, occurs any time one’s velocity is changing, during space shuttle takeoff, for example. Although forces are involved in both cases, the two phenomena appear entirely distinct from each other. One occurs due to the presence of a large mass, while the other occurs due to changes in motion. Now consider an astronaut in a space shuttle with no windows. What could she conclude if she woke up to find a force pressing her into her seat? She might be at rest on the launchpad, in Earth’s gravitational field, or she might be in deep space experiencing no gravitational force but accelerating at a constant rate. The two instances appear very different but the astronaut’s experience of them is identical. This is the equivalence principle. There seems to be an equivalence principle in learning with multimedia, albeit much less profound, which parallels the equivalence principle of General Relativ- ity. Although others have expressed similar ideas about the interchangeability of multimedia, I apply the term equivalence principle to emphasize similarities with its physics counterpart. Consider a student reading a book with words and pictures about Newtonian mechanics. Then, consider the same student watching a movie about Newtonian mechanics. The two experiences appear very different. One involves written text and static images while the other involves spoken text and dynamic images. If we found that following these two instructional treatments, our hypothetical stu- dent performed equally well on the same test, what could we conclude about the two different forms of multimedia? We might, like Lewis (1995), suspect that a movie may not be an appropriate medium for teaching Newtonian mechanics, or perhaps that the movie was deficient in content or presentation. The alternative is to conclude that both media encouraged similar cognitive processes in the student. The equivalence principle in multimedia then states that the relevant cognitive pro- cesses inspired by different formats of multimedia can be made indistinguishable, 12by choosing appropriate methods. Both equivalence principles, like any new ways of looking at the world, unravel the alleged paradoxes of previous research. Although experiments in search of the aether were undoubtedly useful in developing our understanding of space-time, ac- cepting that there is no privelaged reference frame made repeated precise measure- ments seem unnecessary. Similarly, experiments searching for ‘media effects,’ were always doomed to failure by the multimedia principle of equivalence. Both equivalence principles change the way their related phenomena are viewed and illuminate critical areas for consideration. In General Relativity, matter, it was realized, warped space-time so the distribution of matter in the universe and the geometry of space became central concerns. With the multimedia equivalence prin- ciple, the cognitive processes necessary for learning and methods by which they can be triggered become central areas of investigation. Furthermore, the multime- dia equivalence principle implies that teaching and learning techniques developed in different forms of multimedia learning can be applied with similar successes across platforms. 1.4 Conclusion Since its invention, multimedia has become increasingly sophisticated and acces- sible. Most recently, computer technology has allowed for the creation and prop- agation of multimedia with increasing speed. The development of the technology itself has far outpaced efforts at understanding how people learn with multimedia (Rieber 1990). Excitement and intuition displaced research and critical thinking at the outset of each new educational technology. Comparative media studies sought but failed to find evidence of media effects. This line of reasoning obscured the need for theories that explain the interaction between learner and multimedia and how it gives rise to productive cognitive activities for learning. The urge to intro- duce technologies into schools limited the research that was done and lent an air of maturity to the technologies, discouraging further research. Finally, with shifting perspectives of the teaching and learning process, establishing a coherent base of 13theory was next to impossible. Despite the lack of reliable supporting research, multimedia technologies have become commonplace in educational establishments. The costs of technology have dropped dramatically such that virtually every student has access to a computer (Cuban, Kirkpatrick & Peck 2001), where, during the introduction of film, schools had at most one projector. Sophisticated multimedia has also become a larger part of students’ lives with films, television, and the Internet accounting for much of their entertainment and education. Students rely on computers to produce reports and on the Internet to access virtually limitless amounts of information instantly. Tech- nology failed to live up to the promises of its promoters, but it has permeated the school system quite independently of the work of researchers. Without appropriate supporting research however, the successes of multimedia are bound to be unpre- dictable (Sweller 2004). We are doomed to invest significant amounts of money, time and effort in developing multimedia resources that fail to promote meaningful learning. The problem can be refined in terms of the questions considered by multimedia developers. For example, when are interactive resources advantageous over non- interactive media? How does one handle, if at all, the topic of misconceptions? Should the material be presented by a single speaker as in a lecture? Should this speaker appear on-screen or provide narration only? Are different methods advan- tageous for novice and expert learners? Should interesting examples be included to keep the viewer’s attention if only of tangential relevance? Although the possi- ble questions of this type are endless, there are clearly a handful that are vital for understanding and developing effective multimedia. The equivalence principle, that all forms of multimedia can be made equally ef- fective, yields three major implications for this study. First, it warns against search- ing for differences in learning simply due to the use of different media, an enterprise that has a long history of failure. Second, and more informatively, it suggests that teaching and learning experiences that have proved effective in general educational studies can be recreated in multimedia. If these experiences are inherently rare or difficult to facilitate, multimedia can act as a substitute. Third, the equivalence prin- 14ciple underscores the strong link required between sound theory and experiment. In order for a learning experience to be uniquely beneficial, there must be strong theo- retical support for the mechanisms proposed and the anticipated results. Similarly, the results must be capable of informing theory to move both design and research forward. For media proponents, acceptance of the equivalence of learning experiences with different technologies, from textbooks to lectures to computers, would elimi- nate the need for further research; however in this thesis, it is this very equivalence that forms the jumping-off point. Decades of unproductive comparative media stud- ies have left unanswered the questions relating to which methods can be employed in multimedia to achieve the greatest conceptual learning gains. Although the focus of the research is on establishing principles for effective multimedia design, the implications of the research should be generalizable across a range of environments in which students learn from words and pictures. Virtually any presentation that can be created in a classroom or lecture setting can be recreated as multimedia. In addition, the stable and reusable nature of multimedia makes it an ideal tool for carefully investigating methodological differences in teaching. 1.5 Advance organizer The main finding of my thesis research is that multimedia which involves explicit discussion of alternative conceptions is more effective for learning than more con- cise expository summaries. This was demonstrated three times in two different areas of physics with students with different levels of prior knowledge. Supporting data from an empirical study on quantum mechanics are located in Chapter 8 and data from two Newtonian mechanics studies are reported in Chapters 9 and 10. Students were better able to learn with misconception-based multimedia, in which they also invested more mental effort. The construct of mental effort was initially introduced in Chapter 3 as it relates to cognitive load theory and it was measured directly in Chapter 10. The three multimedia learning studies were informed by two main bodies of 15theory, constructivism (Chapter 7) and cognitive theories of learning (Chapter 3). Three preliminary studies also helped identify the research questions and methods. Multimedia learning theories were investigated in their applicability to authentic classroom practice (Chapter 4). Quantum mechanics teaching (Chapter 5) and re- sulting learning (Chapter 6) were explored to understand the challenges and oppor- tunities presented by physics education in the local context. 16Chapter 2 Methodology As outlined in Chapter 1, theories of multimedia design and use are still developing and thus far it has been difficult to bridge the gap between research and practice. Therefore the ‘design experiment’ methodology was selected for this investigation. This research method is becoming increasingly accepted especially in the study of teaching interventions in authentic settings (Lagemann & Shulman 1999, Klahr & Li 2005). The methodology makes use of numerous different data collection and analysis techniques and iterative cycles of design, development, implementation, analysis, and redesign. The underlying goal of design experiment research is to build upon theory while developing effective interventions in authentic contexts. In this chapter I outline the design experiment methodology, its origins, charac- teristics, strengths, and weaknesses. I discuss how the methodology was applied in this investigation including how challenges were addressed. In the process, I present the layout of the thesis, indicating the questions asked and conclusions drawn at each stage of the iterative research process. 2.1 Design experiments Why use design experiments? A common criticism of educational research is that it fails to translate effectively into improved practice. Some fault the lack of rigorous scientific methods in educational research (National Research Council 2002), while 17