Educational modelling language: modelling reusable, interoperable, rich and personalised units of learning

Nowadays there is a huge demand for flexible, independent learning without the constraints of time and place. Various trends in the field of education and training are the bases for the development of new technologies for education. This article describes the development of a learning technology specification, which supports these new demands for learning challenging the new technological possibilities. This specification is named Educational Modelling Language (EML) and is developed by the Open University of the Netherlands. p. 1

A unit of learning is an artefact that is designed for learners to achieve one or more interrelated learning objectives. A unit of learning cannot be broken down into its component parts without losing its semantic and pragmatic meaning and its effectiveness towards the attainment of the learning objectives. In practice you see units of learning of all types: courses; study programmes; workshops; practicals; lessons, etc, are all examples of units of learning. EML is defined as a semantically rich information model and binding, describing the content and process within units of learning from a pedagogical perspective in order to support reuse and interoperability (see Koper, 1991, 1998, 2000). To state it differently: EML is a semantic notation for units of learning to be used in e-learning. p. 1

We define learning technology specifications as specifications of methods and techniques which support the realisation of e-learning. The design and development of learning technology specifications is a global issue. Several initiatives at a national or regional level as well as industrial consortia (IMS Global Learning Consortium) and expert based initiatives (IEEE, LTSC and ADL, Advanced Distributed Learning) are developing learning technology specifications. p. 2

Until now the focus of developing learning technology specifications has been on devel- oping specifications for learning objects. A learning object is defined by the IEEE LTSC (2000) as any entity, digital or non digital, that can be used, reused or referenced during technology supported learning, eg, exercises, cases, study tasks, etc. The specifications for learning objects have primarily been designed to ensure interoperability, focusing on technology issues and reuse of learning objects. The instructional value of learning objects is barely discussed. p. 2

Another major problem with learning objects at this moment is that they are not typed to their usage in the context of a unit of learning. There is a lack of semantic relationship between different types of learning objects in the context and use in an educational p. 2

setting (Koper, 2001b). Another problem is that there are from our point of view a variety of learning objects, such as learning activities, knowledge objects, questionnaire objects, but in the learning object specifications they are all typed as learning objects. p. 3

This reasoning led to the conclusion that we must focus on the development of a learning technology specification which provides a pedagogical framework of different types of learning objects, expresses the relationships between the typed learning objects and defines the structure for the content and behaviour of the different learning objects. This specification must also support objectivist and constructivist views of learning. In order to attain the objectives described above we started the project to develop EML in 1998. In the next paragraphs the focus is on the development of EML. Eventually, EML is presented and the steps to be taken in order to implement a pedagogical design in EML are described. p. 3

The development process of EML consisted of separate iterations of analysis, design, implementation, test and evaluation. The complete development process of EML took about three years and was conducted by a large variety of experts such as educational technologists, ICT-experts, XML-experts, etc. In this section the method of development is described. p. 3

A. General requirements • EML should describe a model for a unit of learning. • EML should describe units of learning in a formal way, so that automatic processing is possible. This includes: editing, storage, assembly and delivery. • EML should use an interoperable notation for units of learning. Through this, invest- ments in educational development will become resistant to technical changes and conversion problems. • EML should describe the units of learning so that repeated execution is possible. This means that EML should model artefacts that are designed and developed in advance and not the artefacts that are produced in runtime. • EML should model all the content resources and communication services, which are present in the unit of learning. • EML should not describe the actual ‘run’ of a unit of learning for actual learners at a given time, but instead it must describe the general case which can be instantiated as many times as necessary for different learners at different times. • EML should allow the packaging of a unit of learning in one container or file to enable transportation. However, it must also be possible to break the container down to its p. 3

subcomponents or to edit subcomponents and integrate them into an unit of learning by reference. • EML should describe metadata for the unit of learning and all of its reusable sub artefacts in order to identify the characteristics and ownership, to support search, reference and assembly. • EML should be built on available standards and specifications where possible. This includes specifications from IMS (http://www.imsproject.org), IEEE LTSC (http://www.ltsc.ieee.org/), ISO/IEC JTC1/SC36 (http://jtc1sc36.org/), IACC (http://www.aicc.org), and ADL SCORM (http://www.adlnet.org). • EML should make it possible to produce, mutate, preserve, distribute and archive units of learning and all of its containing learning artefacts. p. 4

B. Instructional design requirements for units of learning • EML should be able to fully describe a unit of learning, including all the typed learning objects, the relationship between the objects and the activities and the workflow of all students and staff members with the learning objects, regardless of whether these aspects are represented digitally or non-digitally. • EML should define the conditions under which different learning artefacts can be aggregated into a valid unit of learning. • EML should explicitly express the semantic meaning of the different learning artefacts within a unit of learning, using a pedagogical vocabulary from the educational domain. • EML should allow users to map the pedagogical terminology used in EML to their own terminology. • EML should allow the modelling of different kinds of pedagogical models, includ- ing the more traditional teacher directed and information transmission based models, as well as the more student centred, collaborative and constructivist approaches. • EML should make a distinction between different roles, especially learner and staff roles. However, it should not be rigid in allowing certain kinds of activities only for certain roles. One must be able to assign all kinds of activities to staff as well as to learner roles in order to be able to shift learning functions from the one to the other (Shuell, 1988; Koper, 1995). • EML should enable the definition of formal criteria for a student to meet in order to complete (parts of) a unit of learning. This means that assessment procedures and tools, along with other completion facilities must be available. In this respect, classical testing such as multiple-choice testing, as well as new assessment models such as performance tests or portfolio assessment should be supported (Hambleton, 1996; Sluijsmans, 2002). • EML should be able to describe personalisation aspects within units of learning, so that the content and activities within units of learning can be adapted based on the preferences, prior knowledge, educational needs and situational circumstances of users. • EML should be able to use and define properties in a learner dossier, in order to build portfolios, support monitoring facilities and support student tracking. p. 4

• EML should allow units of learning to contain other units of learning. This allows the building of a curriculum (a unit of learning) from underlying courses (a unit of learning) which itself can consist of different units of learning (eg, a lesson). Also in the analysis stage, vocabulary and preliminary models and architectures for EML were defined. p. 5

In order to make EML more generic, a new analysis was conducted. The idea was to define a pedagogical meta model, which was neutral to the different approaches to learning and instruction, and then take the entities of the meta model as a starting point for modelling EML (meta model is elaborated in next paragraph). This resulted in the second implementation of EML 0.5. p. 5

The result of the development process was the release of EML 1.0. As written above, the focus in this paragraph is on the models behind EML and the structure of EML. The Reference Manual, which provides a full description of all the elements and attributes within EML and the DTD as well, can be downloaded at http://eml.ou.nl. In this paragraph we make a distinction between: • the pedagogical meta model behind EML; • the unit of learning model; • basic structure of EML. p. 6

During the development of EML a pedagogical meta model emerged, which is the base for EML. A pedagogical meta model is a model which models pedagogical models. This means that pedagogical models could be described (or derived) in terms of the meta model. The pedagogical meta model is based upon educational research, especially in the field of learning psychology and instructional design (Koper, 2001b). p. 6

Greeno, Collins and Resnick (1996) make—in a meta-analysis—a distinction between three major streams of instructional theories: 1. empiricist (behaviourist); 2. rationalist (cognitivist and constructivist); 3. pragmatist-sociohistoric (situationalist). p. 6

These instructional theories have different views on topics such as: knowledge, learn- ing, transfer and motivation. Based upon these theories there are hundreds or more theoretical or practical theories and models of learning and instruction, for instance competency based learning, project based learning, mastery learning, problem based learning, case based learning, etc. Most of these models were studied and analysed. The commonalities were mapped and the differences listed in order to derive the pedagogical meta model. p. 6

This educational research resulted in the five axioms of the pedagogical meta model: 1. A person learns by performing goal directed activities in an environment. 2. When a person has learned, he is able (a) to perform new activities or perform activities better or faster in similar environments or (b) to perform the same activities in different environments. p. 6

3. An environment consists of a set of objects and/or human beings that are related in a particular way. 4. A person can be encouraged to perform certain activities when: a. b. The activities can be performed by this person, given the requirements in terms of pre-knowledge, personal circumstances and the performance context. The required environment is made available. The person is motivated to perform the activities. c. 5. What had been posed here with respect to a single person, also applies to a group of persons. p. 7

instruction has been defined as follows: ‘Instruction is a process which aims at accomplishing and measuring learning results’. p. 7

It can be concluded from the axioms that instruction should consist of providing stu- dents with coherent series of activities, including specific learning environments, so that learning actually can take place. Assessment of what has been learnt may consist of providing students with specific activities, which enable them to show that the aimed learning objectives have been obtained. p. 7

The core concept of the unit of learning model, as expressed in Figure 1, is that, regard- less of pedagogical approach, a person gets a role in the teaching-learning process, typically a learner or a staff role. In this role he or she works towards certain outcomes by performing more or less structured learning and/or support activities within an envi- ronment. The environment consists of the appropriate learning objects and services to be used during the performance of the activities. Which role gets which activities at what moment in the process is determined by the method or by a notification. p. 7

The method is designed to meet learning objectives and presupposes certain prerequisites. The method consists of one or more concurrent play(s); a play consists of one or more sequential act(s). A method may contain conditions, ie, If-Then-Else rules further refine the visibility of activities and environment entities for persons and roles, by defining Boolean expressions on their properties. A notification is triggered by an outcome and can make a new activity available for a role to perform. Activities can be assembled into activity-structures. A structure can model a sequence or a selection of activities. In a sequence, a role has to complete the different activities in the structure in the order provided. In a selection, a role may select a given number of activities from the set provided in the activity-structure. p. 7

Environments can contain two basic types: 1. Located learning objects. In EML the learning objects are classified as the following types: knowledge-objects, tool-objects and test-objects. p. 7

2. Services. A service relates to a concrete service facility available at runtime. Examples of a Service include a discussion forum, chat rooms, monitoring tools, search facil- ities, etcetera. p. 8

The first step in designing with EML is to specify who plays a role in the instructional design. A separation has been made between student roles and staff roles. Which roles are present in the EML-design depends on the chosen pedagogical model. p. 9

Activities The second step in the design is to specify what persons in these roles are expected to do. In EML this is referred to as ‘Activities’. There are two types of activities: learning activities, to be performed by student roles, and support activities, to be conducted by either staff or student roles. p. 10

A full 440-hour course delivered in mixed face-to-face and distance teaching mode was developed in EML and run with actual students (instead of the regular course which was provided before). This course was part of a curriculum of an institute for higher vocational education in the field of Hotel Management. The average number of students who attended the four presentations of this course was between fifty and sixty. Also a complete competency based curriculum was developed for dual mode education, consisting of six courses. p. 13

Eight distance teaching courses in several fields, public administration, psychology, law, methodology, with thousands of pages of domain specific content in EML and a variety of instructional models (traditional and more competency based models as well) were developed and delivered to students of the Open University of the Netherlands. • Several units of learning were developed for testing and demonstrating issues, dealing with the more complex constructs of EML, such as personalisation and workflow (learning flow) modelling. p. 13

To broaden the use of EML, we utilised the learning technology specifications of different international bodies, such as the IMS Global Learning Consortium. Within IMS, EML was selected as the basis for the IMS Learning Design specification (LD), which was approved as an official IMS specification in February 2003 (IMS Learning Design Spec- ification is available at http://www.imsproject.org/learningdesign/index.cfm). IMS LD and EML share the same philosophy, aim and model, as presented in this article. p. 14

Brown, J. S., Collins, A. & Duguid, P. (1989). Situated cognition and the culture of learning. Educational Researcher, 18, 1, 32–42. Dillenbourg, P. & Schneider, D. (1995). Mediating the mechanisms which make collaborative learning sometimes effective. International Journal of Educational Telecommunications, 1, 131– 146. Forte, E., Wentland-Forte, M. & Duval, E. (1997). The ARIADNE project (part I and II): knowledge pools for computer based and telematics supported classical, open and distance education. European Journal of Engineering Education, 22, 1/2, 61–74 (part I) en 153–166 (part II). Greeno, J. G., Collins, A. M. & Resnick, L. B. (1996). Cognition and learning. In D. C. Berliner & R. C. Calfee (Eds), Handbook of educational psychology (pp. 15–46). New York: Simon & Schuster/ Macmillan. Hambleton, R. K. (1996). Advances in assessment models, methods, and practices. In D. C. Berliner & R. C. Calfee (Eds), Handbook of educational psychology (pp. 899–925). New York: Simon & Schuster/Macmillan. Hermans, H. J. H., Manderveld, J. M. & Vogten, H. F. (2003). Educational modelling language. In W. M. G. Jochems, J. J. van Merriënboer & E. J. R. Koper (Eds), Integrated E-learning (pp. 80–99). London: Kogan Page. IEEE LTSC (2000). Learning Technology Standards Committee. http://ltsc.ieee.org Koper, E. J. R. (1991). Inscript: a courseware specification language. Computers & Education, 16, 2, 185–196. Koper, E. J. R. (1995). PROFIL: a method for the development of multimedia courseware. British Journal of Educational Technology, 26, 2, 94–108. Koper, E. J. R. (1998). Definitiestudie Elektronische Leeromgeving. Heerlen: Open Universiteit Nederland. Koper, E. J. R. (2000). From change to renewal: educational technology foundations of electronic learning environments. Retrieved from http://eml.ou.nl/introduction/docs/koper-inaugural-address.pdf Koper, E. J. R. (2001a). Van verandering naar vernieuwing. In P. Schramade (Ed.), Handboek Effectief Opleiden. ’s Gravenhage: Delwel. Koper, E. J. R. (2001b). Modeling units of study from a pedagogical perspective: the pedagogical meta model behind EML. Retrieved from http://eml.ou.nl/introduction/docs/ped-metamodel.pdf) Leenders, M. R. & Erskine, J. A. (1989). Case research: the case writing process (3rd ed.). Ontario: University of Western Ontario. p. 14

check out Koper 2001b p. 14

Manderveld, J. M. & Koper, E. J. R. (1999). Building a competence based electronic learning environment. In B. Collis & R. Oliver (Eds), Ed-Media 1999: World Conference on Educational Multimedia, Hypermedia & Telecommunnications (pp. 1543–1544). Charlottesville: AACE. Oliver, M. & Bradley, C. (1999). Examples of best practice in the use of multimedia in higher education (EXE report n02). London: University of North London. Robinson, D. G. & Robinson, J. C. (1995). Performance consulting: moving beyond training. San Francisco: Berrett-Koehler. Roblyer, M. D. & Edwards, J. (2000). Integrating educational technology into teaching (2nd ed.). Upper Saddle River: Merrill, Prentice Hall. Schlusmans, K., Slotman, R., Nagtegaal, C. & Kinkhorst, G. (Eds) (1999). Competentiegerichte leeromgevingen (competence based learning environments). Utrecht: Lemma. Scott Grabinger, R. (1996). Rich environments for active learning. In D. H. Jonassen (Ed.), Handbook of research for educational communications and technology (pp. 665–692). New York: Macmillan. Shuell, Th. J. (1988). The role of the student in learning from instruction. Contemporary Educa- tional Psychology, 13, 276–295. Sluijsmans, D. (2002). Student involvement in assessment: the training of peer assessment skills, doctoral dissertation. Heerlen: Open Universiteit Nederland. Spencer, L. M. & Spencer, S. M. (1993). Competence at work: models for superior performance. New York: John Wiley & Sons, Inc. Stacey, P. (2003). People to people, not just people to content. Presentation at the IMS Open Tech- nical Forum, Vancouver. Retrieved from http://www.bctechnology.com/statics/pstacey-feb1403.html) Wiley, D. A. (2002). The instructional use of learning objects. Bloomington: Agency for Instructional Technology. p. 15