Microsoft Word - elliott.doc Australasian Journal of Educational Technology 2008, 24(2), 137-142 Quantifying the reuse of learning objects Kristine Elliott and Kevin Sweeney The University of Melbourne An Outstanding Paper Award recipient, ascilite Singapore 2007 Conference This paper reports the findings of one case study from a larger project, which aims to quantify the claimed efficiencies of reusing learning objects to develop e-learning resources. The case study describes how an online inquiry project Diabetes: A waste of energy was developed by searching for, evaluating, modifying and then integrating as many pre-existing learning objects as possible into a learning design. Development times for the reuse approach were recorded and compared to estimates for the de novo development of an equivalent project. Outcomes suggest that considerable savings can be made using the reuse approach; we estimate a threefold increase in time to develop the Diabetes project using new objects in comparison to reuse. In this case study, gaining permission from owners to reuse objects was not a barrier to reuse. However, in some circumstances, being unable to source pre-existing objects to meet specific requirements, or having to modify objects for reuse, could be problematic. Introduction One of the major developments in e-learning this decade has been the notion of reuse of digital resources. This approach sees resources created for one particular learning context, made available for reuse in another context. To be available, resources should be searchable and accessible through the Internet (Koper, 2003). Having been derived from object oriented models of software programming, the basic component of the reuse approach is the learning object; small, discrete blocks of educational content that can stand alone, be aggregated to form more complex learning objects, and/or sequenced to create novel learning experiences. In this paper, we use Wiley’s (2002) definition of a learning object as “any digital resource that can be reused to support learning”. Many benefits are expected from the reuse approach, including improved economic and time efficiencies for resource development, and more effective learning and teaching practices. However, the potential of the learning object economy is yet to be realised. Practical difficulties that educators encounter while sourcing, adapting, aggregating and sequencing learning objects (Koper, 2003), along with socio-cultural attitudes towards collaboration and sharing resources have been identified as potential barriers to the reuse approach (Littlejohn, 2005). Therefore, it is not entirely clear at present whether it is feasible to expect educators to search for, re-purpose and integrate existing learning objects into new teaching contexts, in order to create effective teaching experiences. In 2004, we began a project to create an online learning environment to enhance the learning and teaching of biochemistry for second year students undertaking 138 Australasian Journal of Educational Technology, 2008, 24(2) compulsory biochemistry units in their courses of Bachelor of Pharmacy and Bachelor of Pharmaceutical Sciences, at a major Australian university. The resulting learning environment contained a series of inquiry projects that covered a range of topics and incorporated learner tasks and resources. The rationale behind the development was to provide a flexible learning tool that focussed on biochemistry principles and concepts that traditionally students had found difficult to grasp. More generally, the inquiry projects were designed to promote understanding of the scientific method of inquiry and development of life long learning strategies. The development phase of this project was used as an opportunity to investigate practical aspects of the learning object approach. Rather than developing the inquiry projects de novo, they were created by searching for, aggregating and sequencing as much pre-existing material as possible. The main objective of the study was to quantify the claimed efficiencies of the reuse approach. To do this, the time taken to develop projects using the reuse approach was documented and then compared to time estimates for the de novo development of equivalent projects. Furthermore, we sought to identify the types of problems that educators might encounter during the reuse approach. While reuse data was collected during development of all the inquiry projects contained within the learning environment, this paper presents the findings of one case study from the larger project; the development of the online inquiry project Diabetes: A waste of energy. Method The process used to develop Diabetes: A waste of energy by reusing learning objects (LOs) is described below. It is based on Koper’s (2003) top down approach that begins with a learning design that LOs are searched for, and then integrated into. Learning design The first phase involved the development of a learning design, which specified tasks students were required to perform, the resources and feedback supporting student activities, and the sequence in which these events occurred. This phase was an iterative process with gradual refinement of the design over time. While an in depth description of the learning design is not the focus of this paper, it can essentially be described as a template for inquiry projects in which students assume the role of professional scientist to solve real life problems. The design incorporated a standardised scientific inquiry process that students were guided through in order to solve these problems. Theories of inquiry learning (de Jong, 2006), problem based learning (Barrows & Tamblyn, 1980), and constructivist philosophy (Mayes & de Freitas, 2004) informed the learning design. For further details about the learning design see Elliott, Sweeney and Irving (in press). Learning object requirements With a clear understanding of the tasks, resources and supports composing the learning design, the next phase was to broadly define the LOs required to develop a project that addressed the biochemistry principles of bioenergetics and metabolism. The real life problem of diabetes was an ideal way of introducing these principles to students (e.g. “Guinevere, a type 1 diabetic, has collapsed at a party – investigate Elliott and Sweeney 139 details of metabolism to discover why”). Therefore, LOs were required that would set the context of the problem scenario, present the medical signs and symptoms of diabetes, explain the biochemistry of normal and starvation metabolism, illustrate the pathways involved in carbohydrate and lipid metabolism, indicate the sites of metabolic pathways, demonstrate the role of insulin and what goes wrong in diabetes, and show how diabetes is treated. LOs could also be used to allow students to reflect on and test their understanding of the content. Search To find LOs, repositories such as ARIADNE, BIOME, Bitstream’s medical links, CAREO, European Schoolnet, HEAL, LEARNet, MERLOT, TryScience and UCEL were searched, followed by use of the internet search engine Google. Returned LOs were evaluated for their pedagogical value (i.e. did their content match the learning objectives of the project), technical suitability and compatibility with the interface design. Some latitude was allowed during evaluation; objects that were not quite what we had sought but where still suitable with minor modification were accepted. Owners of selected LOs were contacted to obtain permission to reuse their materials. While the time to obtain permission was not specifically recorded for each LO, it was our experience that on average this process took 0.5 hr per LO. If existing objects could not be found that met specific content requirements, then new objects were produced. The time taken to search for, evaluate, and create all LOs used in the project was recorded. Production LOs identified for use were embedded into HTML pages, along with original content created by content experts. LOs requiring technical or educational changes were modified at this stage. Generally, the Diabetes project consisted of HTML pages containing images in GIF or JPEG format, Macromedia Flash animations and interactive exercises programmed in Javascript. In terms of the larger study outlined in the Introduction, all projects were incorporated into the empty shell of HTML templates used to create The Virtual Laboratory, a pre-existing learning environment developed by St Vincent’s Institute of Medical Research and the Victorian Department of Education (Brack et al., 2003). This shell was chosen because its structure and function suited our purposes and saved us from having to create another learning environment. A license was obtained to use and modify The Virtual Laboratory shell. A professional multimedia development unit dedicated to the development of educational technologies carried out production. The time spent modifying and embedding all reused LOs into HTML pages was recorded. De novo development An estimate was made of the time it would take to develop an equivalent project by creating all LOs anew, instead of reusing materials. To do this, an experienced web developer and educational designer together reviewed all the LOs in the Diabetes project and estimated the time it would take to develop them anew. Estimates were based on real development times documented for the production of similar materials by the multimedia development unit. To avoid possible underestimation, a further 20% was added to estimates, resulting in an estimated time range. Time for tasks such as educational design and writing original content were not included in the estimate as these tasks were common to both the de novo and reuse methods of development. 140 Australasian Journal of Educational Technology, 2008, 24(2) Results and discussion Learning object use in the project Diabetes: A waste of energy is summarised in Table 1. Searches returned a total of 77 LOs that were broadly consistent with our content requirements. Following evaluation, 38 (49%) of these were selected for reuse. Obtaining permission from owners to use their LOs was not an obstacle to reuse. Most owners were willing to share their materials for non-commercial, education use, particularly if appropriate acknowledgement was made. In this case study, permission to reuse was rejected for one LO because the owner requested commercial terms of use that were unsuitable to us. Two new LOs were created because of specific content needs; a schematic diagram showing integration of all the metabolic pathways involved in diabetes, and a close up image of a person monitoring blood glucose levels. Therefore, the final proportion of reused LOs in the project was 95%. Table 1: Learning object use in the project Diabetes: A waste of energy Permission to reuseReturned from search Selected Rejected Granted New Total 77 38 1 37 2 39 The types of LOs reused in the development of the Diabetes project are shown in Table 2. By far, images were the major type of LO reused (74%). In this case study, images were used to set the context of the problem scenario, demonstrate metabolic pathways and chemical structures, indicate sites of metabolic pathways, and graphically illustrate hormone effects. Table 2: Type of learning object used in the project Diabetes: A waste of energy Image Interactive image Animation Tutorial Text Total 29 (74%) 1 (3%) 5 (13%) 2 (5%) 2 (5%) 39 (100%) Of the 37 LOs reused, 36 (97%) required some form of modification to enhance their educational value. This type of modification included addition of introductory text, title, explanatory text, labels or arrows, and was primarily carried out to contextualise the LO within the inquiry scenario. Technical modifications such as changing file formats to one more suitable for the web (e.g. from TIFF to PNG), were made to 22 (60%) LOs. Development times for Diabetes: A waste of energy are shown in Table 3. It took 14 hrs to search for, evaluate and create the 39 LOs in the Diabetes project. The time taken to obtain permission to reuse LOs was 19 hrs. The time taken to modify all reused LOs was 15 hrs, which included 8 hrs for technical modifications and 7 hrs for educational modifications. Therefore, in total these tasks took 48 hrs to complete (see Table 3). In comparison, we estimated that an equivalent project developed by creating all LOs anew, would be in the range of 133-160 hrs to develop (see Table 4). Table 3: Development times (hrs) for the reuse method for the project Diabetes: A waste of energy ModificationSearch Permission Contexual Technical Total 14 (29%) 19 (40%) 7 (15%) 8 (16%) 48 (100%) Elliott and Sweeney 141 Table 4: Development time for the project Diabetes: A waste of energy Reuse (actual) De novo (estimate range) 48 (hrs) 133-160 (hrs) Conclusions In the context of developing the online inquiry project Diabetes: A waste of energy by reusing LOs, it was not difficult to find LOs that were broadly consistent with our needs. However, only 49% of returned LOs met our specific requirements. The search and evaluate process that was used in this study did not always return suitable LOs the first time around and needed to be repeated. This meant that many more LOs were returned from searches, than were actually used in the final product. The search and evaluate process accounted for 29% of development time. Most LO owners willingly gave permission to reuse their resources for the non- commercial, educational purpose of the inquiry project. However, the time taken to contact and negotiate permissions from owners accounted for 40% of development time. Many owners made specific requests, for example, as to how copyright statements should be worded, or to be given access to the learning environment to see how their resource(s) was being used. The final proportion of reused LOs in the Diabetes project was 95%. There were only two instances when pre-existing LOs could not be found to meet the evaluation criteria. In this case study, the majority of LOs reused were classed as raw assets (74%), containing no inherent educational context (Koppi et al., 2000). We found that it was necessary to modify 97% of the pre-existing LOs to match the context of the inquiry scenario. The time taken to recontexualise LOs accounted for approximately 15% of development time. Furthermore, technical modifications needed to be made to 60% of LOs, and accounted for 16% of development time. In total, modifications accounted for 31% of development time. A comparison of the time taken to develop the inquiry project by reusing learning objects, with an estimated range of time it would take for the de novo development of an equivalent project, suggests that considerable savings can be made by reusing LOs. We estimate that it would take a threefold increase in time to develop the Diabetes project with newly created LOs as compared to the reuse approach. In this study, the low proportion of LOs returned from searches that were actually reused (49%) would have reduced the efficiency of the reuse approach. Therefore, there is potential for greater savings as the adoption of e-learning catalogues and repositories become more mainstream and searching for LOs becomes more efficient. While acknowledging that production using the reuse approach was carried out by a professional multimedia unit with experienced and skilled staff, and that development times determined for the de novo approach were estimates based on actual times, this study has provided an indication of the type of savings that educators can expect by using the reuse approach. Moreover, it has identified certain circumstances (e.g. being unable to source a pre-existing LO to meet specific requirements, having to modify LOs for reuse) that may be problematic for some educators working without the support of a professional multimedia unit. Since this article was accepted for publication in Proceedings ascilite Singapore 2007, we have continued to analyse the reuse data from other inquiry projects developed for the 142 Australasian Journal of Educational Technology, 2008, 24(2) online learning environment. These results reveal that time efficiencies vary with the nature of the project. While it appears that efficiency of the reuse approach is dependent upon the subject matter and complexity of the LOs being sourced, future research is required to clarify the influence of these factors. Acknowledgements The authors would like to thank Dr Helen Irving, Monash University and production staff from the Biomedical Multimedia Unit, The University of Melbourne for their involvement in the development of the program that formed the basis of this research. References Brack, C., Elliott, K. A., Fisher, J. & Stapleton, D. (2003). The Virtual Laboratory: An online program to integrate authentic activities into the biology curriculum. In Interact, Integrate, Impact: Proceedings ASCILITE 2003 (pp. 581-584). Adelaide, 7-10 December. http://ascilite.org.au/conferences/adelaide03/docs/pdf/581.pdf Barrows, H. S. & Tamblyn, R. M. (1980). Problem-based learning: An approach to medical education. New York: Springer. Elliott, K. A., Sweeney, K. & Irving, H. R. (in press). A learning design to teach scientific inquiry. In L. Lockyer, S. Bennett, S. Agostinho & B. Harper (Eds), Handbook of research on learning design and learning objects: Issues, applications and technologies. Idea Group Inc. de Jong, T. (2006). Technological advances in inquiry learning. Science, 312, 532-533. Koper, E. J. R. (2003). Combining reusable learning resources and services to pedagogical purposeful units of learning. In A. Littlejohn (Ed), Reusing online resources: A sustainable approach to eLearning (pp. 46-59). London: Kogan Page. Koppi, T., Hodgson, L. & Bayly, J. (2000). The often missed but essential component for online learning: A learning resource catalogue. In J. Bourdeau & R. Heller (Eds), Proceedings of 12th World Conference on Educational Multimedia, Hypermedia & Telecommunications. Montreal, Quebec, Canada, June 26-July 1, 2000. Littlejohn, A. (2005). Community dimensions of learning object repositories. In H. Gross (Ed), Balance, fidelity, mobility: Maintaining the momentum? Proceedings ASCILITE 2005 (p3). http://www.ascilite.org.au/conferences/brisbane05/blogs/proceedings/KA3_Littlejohn.pdf Mayes, T. & de Freitas, S. (2004). Review of e-Learning theories, frameworks and models. JISC e- Learning Models Desk Study. [viewed 26 Feb 2007] http://www.jisc.ac.uk/ uploaded_documents/Stage%202%20Learning%20Models%20(Version%201).pdf Wiley, D. A. (2002). The instructional use of learning objects. [viewed 20 Aug 2007] http://www.reusability.org/read/ This article received an Outstanding Paper Award at ascilite Singapore 2007 Conference, gaining the additional recognition of publication of a revised version in AJET. The reference for the Conference version is: Elliott, K. & Sweeney, K. (2007). Quantifying the reuse of learning objects. In ICT: Providing choices for learners and learning. Proceedings ascilite Singapore 2007. http://www.ascilite.org.au/conferences/singapore07/procs/elliott-k.pdf Kristine Elliott and Kevin Sweeney are both Research Fellows in educational technology at the Biomedical Multimedia Unit, Faculty of Medicine, Dentistry and Health Sciences, University of Melbourne, 3010 Vic, Australia. Email: kaelli@unimelb.edu.au, sweeneyk@unimelb.edu.au