Pedagogical Design Patterns for E-Learning: A New Approach to Instructional Design Theory
Pedagogical Design Patterns for E-Learning: A New Approach to Instructional Design Theory Helmut M. Niegemann Center for Research on Learning and Instruction University of Erfurt, Germany [email protected]
Silvia Hessel Center for Research on Learning and Instruction University of Erfurt, Germany [email protected]
Steffi Domagk Center for Research on Learning and Instruction University of Erfurt, Germany [email protected]
Abstract: Pedagogical design patterns represent an idea that has been taken from architecture and software engineering and transferred to e-learning. Analogous to design patterns in architecture and software engineering, a pedagogical design pattern should “describe a problem which occurs over and over and again […] and then describes the core of the solution to that problem, in such a way that you can use this solution a million times over, without ever doing it the same way twice.” (Alexander et al., 1977). The idea is presented and some problems specific to pedagogical patterns are discussed.
Pedagogical Design Patterns The idea behind pedagogical design patterns was derived from the idea behind design patterns in architecture realized about 30 years ago by Christopher Alexander and his co-workers (Alexander et al., 1977). Software engineers had adapted this idea for their field (Gamma et al., 1998) and from there it entered the e-learning domain (Goodyear, 2000). The idea emerges at a time when instructional design (ID) as a discipline is being confronted with a paradox problem: Although the spread of e-learning requires more ID expertise, there is a drop in the acceptance of existing ID models as a whole, including problem-based ID models. On the one hand instructional designers ask for valid design principles, but on the other hand their majority seem not to accept the ID models developed over the last 50 years. Most ID models are like set menus, but ID practitioners seem to prefer to eat “à la carte.” Each design pattern refers to a specific design decision and each pattern should contain a clear description of the context and its conditions for use. Several sets of e-learning and pedagogical design patterns have been published in the World Wide Web by various authors or groups. A repository of e-learning patterns has been generated by the partners of an international project (E-LEN) part-funded by the European Community in 2003 and 2004 (cf. http://www.tisip.no/E-LEN). The partners in this project have developed a framework to describe design patterns as follows (only the most relevant categories are mentioned):
International Conference on Computers in Education 2004 NAME of the pattern ABSTRACT (short description of the pattern) PROBLEM (what kind of problem should be solved by using the pattern) ANALYSIS (what makes the problem to be solved a problem) KNOWN SOLUTIONS (are there alternative solutions?) RESEARCH QUESTIONS (open research questions concerning the pattern) CONTEXT (description of the context/s the solution is applicable to/not applicable to) CONDITIONS/RATIONALE (reasons for the applicability of this solution in the context) DISCUSSION/CONSEQUENCES (consequences of use, side effects, implementation issues) REFERENCES (documentation, URLs) RELATED PATTERNS (alternative patterns, patterns to discriminate…) AUTHOR/S REFERENCES (cf. Rusman, van den Broek, & Ronteltap, 2003) Analogous to architectural patterns, pedagogical patterns should not prescribe any specific way to realize a solution; they should rather provide a scientifically based framework for pedagogical design decisions. Designing a multimedia learning environment involves a cascade of design decisions to be made on different levels (Schnotz et al., 2004): from a “strategic level” (e.g. decisions concerning the kind of learning goals), to the level of decisions concerning the screen layout, the wording of the instructional texts and the color of the background. Examples of pedagogical design patterns on the strategic level could be based on the “basis-models” defined by Oser (Oser & Baeriswyl, 2001). A basis-model describes the kind and sequence of mental operations, which as a necessity have to be stimulated by the learning environment if a specific type of learning goal is to be desired. But the way the mental operations are stimulated is open to various pedagogical methods. To present a sample pattern we chose a brief version of the pattern “Concept Building.” NAME: “Concept Building” ABSTRACT: To foster the building of a concept there are 5 elements. Each element describes an absolutely necessary basic mental operation. The elements must be stimulated in their specific sequence. The way to stimulate the elementary operations is a free choice. PROBLEM: The pattern should be used if the learning goal is to convey a new concept. ANALYSIS: New concepts could not be learned and applied long-term if only a definition is given or if several instances are presented. SOLUTION: 1. Direct or indirect stimulation of the awareness in what the learner lready knows regarding the new concept. 2. Introduction of and the working through of a prototype as a valid example of the new concept. 3. Analysis of essential categories and principles that define the new concept (positive and negative distinctions). 4. Confrontation with the new concept in different contexts (application, synthesis, and analysis). 5. Application of the new concept in different contexts (incorporation of different but similar concepts into a more complex knowledge system). RESEARCH QUESTIONS: It is not clear which role the learner's motivation plays.
Pedagogical Design Patterns for E-Learning: A New Approach to Instructional Design Theory CONTEXT:
CONSEQUENCES: REFERENCES: RELATED PATTERNS: AUTHOR/S:
From the position of Oser's instructional design theory, the pattern should be used in all situations where “concept building” is a type of learning objective. The process of building new concepts has been thoroughly studied over several decades. The pattern integrates the results of educational and developmental research. The use of the pattern requires a higher level of knowledge in cognitive/developmental psychology. cf. Oser & Baeriswyl (2001) Pattern for KNOWLEDGE BUILDING (learning the meaning of words); pattern for PROBLEM SOLVING. Pattern proposed by H. Niegemann.
Although the sample design pattern is actually well founded, the quality of pedagogical design patterns in general is in question. It seems for Alexander et al. (1977) that that concern was not as relevant. Their 253 architectural design patterns are highly plausible, something not so strange regarding a domain that is so rich in tradition like architecture. As for design patterns in software engineering, the aspect of quality is being discussed but arguments do not question the quality of a pattern per se, the quality criteria are generally rather clear. In pedagogy things are not as clear. In fact some frequently used design patterns in actuality do not function well: for example there are animations on the screen explained by written texts; there are written texts accompanied by the same spoken text; there are “stories” inserted in e-learning modules following the assumption that interesting stories are more effective in learning motivation (ref. Clark & Mayer, 2003). As replicated experiments have shown, none of the “patterns” mentioned were efficient. So it seems clear that design patterns in the pedagogical domain are not always suitable solutions to instructional design problems. Contrary to architecture a pattern that "describes a problem which occurs over and over and again in our environment, and then describes the core of the solution to that problem, in such a way that you can use this solution a million times over, without ever doing it the same way twice.” (Alexander et al., 1977, x) has to be
questioned. Pedagogical design patterns need thorough proof of their quality; they have to be evaluated. So the question of how to design the evaluation of pedagogical design patterns remains.
Criteria The quality of a pedagogical design pattern has to be proved on (at least) two levels:
Level I: Evaluating the Pedagogical Design Pattern as a Whole We doubt that every pattern which fulfills most of the simpler criteria meets the gist of the pedagogical design patterns developed by Alexander et al. and adapted for education by Goodyear (2003). Therefore, evaluation should include a global expert rating system concerning that aspect. But there are also some other aspects of a whole pattern that may not be assessed by studying the description categories separately (see Rusman, van den Broek, & Ronteltap, 2003): acceptance by designers/developers;
International Conference on Computers in Education 2004
the degree to which a pattern benefits the work of designers and developers objectively; the degree to which a pattern makes the development of learning environments more efficient (efficiency of development); the degree to which the pattern makes the implementation, organization and administration of a learning environment more efficient (efficiency of implementation/organization/ administration); the degree to which learning is fostered after the pedagogical design pattern has been appropriately realized and applied (learning efficiency); usability of the learning environment; suitability regarding the principles of the psychology of learning and instruction.
Level II: Evaluating the Categories Separately Even if a candidate pattern seems to meet the very idea of pedagogical design patterns, any description of a category can show deficits, which could hinder the pattern receiving acceptance and being widely communicated. Therefore we need criteria for every single category used to describe a pattern. Relevant questions to guide that evaluation have been formulated by Rusman, van den Broek, and Ronteltap (2003).
Methods In principle there are several methods suitable for the evaluation of pedagogical design patterns, but different methods would contribute to the evaluation of different aspects of a pattern. Therefore only one specific method cannot be appropriate for a comprehensive evaluation. Instead we need a group of methods, selected and combined to assess a multitude of aspects regarding the effectiveness and efficiency of every pedagogical design pattern. Structured expert ratings are a most important method, especially in a time when the idea of pedagogical design patterns has not yet been widely spread and as long as no stable community exists cultivating the idea. On the other hand there are also “objective” standards to be fulfilled. Evaluations should assess the acceptance of the pattern by designers and developers as well as the psychological foundations or the evidence base of every pattern. The repertoire of methods referring to users is comprised of: user ratings (designers/developers); the assessment of the actions of designers/developers through self-reports (interview, structured questionnaire); the assessment of the efficiency of the product including observation of acceptance (interviews, questionnaire, group interview with focus groups); recording the costs of development (structured questionnaire); the costs of implementation (structured questionnaire); and the costs of the organization and administration of the pattern (structured questionnaire) as well as assessing the benefits of the use of the pattern in question (interview, structured questionnaire, focus group). Last but not least there should be learners’ ratings (questionnaire) and an assessment of learning outcomes (standardized tests).
Problems and Prospects The evaluation of pedagogical design patterns is a new task; there has been no experience made as of yet. Therefore the catalogue of criteria and methods is provisional as a necessity. It will be developed further: some criteria may prove to be less suitable, and some may need more elaboration. Evaluation processes will uncover gaps in instructional design knowledge and hopefully initiate new research activities.
Pedagogical Design Patterns for E-Learning: A New Approach to Instructional Design Theory
Reference List Alexander, C., Ishikawa, S., Silverstein, M., Jacobson, M., Fiksdahl-King, I., & Angel, S. (1977). A pattern language: towns, buildings, construction. New York: Oxford University Press. Avgeriou, P., Papasalouros, A., & Retalis, S. (2002). Patterns for designing learning managements systems, 2004, from www.tisip.no/E-LEN Avgeriou, P., Papasalouros, A., Retalis, S., & Skordalakis, M. (2003). Towards a pattern language for learning managements systems. Educational Technology & Society, 6(2), 11-24. Clark, R. & Mayer, R. E. (2003). E-learning and the Science of Instruction. San Francisco: Jossey-Bass. Gamma, E., Helm, R., Johnson, R., & Vlissides, J. (1998). Design patterns CD. Elements of reusable object oriented software. New York: Addison-Wesley Longman. Goodyear, P. (2000). Environments for lifelong learning: ergonomics, architecture, and educational design.(pp. 118) In J.M. Spector & T. Anderson (eds.). Integrated and holistic perspectives on learning, instruction & technology. Dordrecht: Kluwer. Goodyear, P. (2003). Towards a pattern language for e-learning: a discussion paper. Unpublished manuscript. Niegemann, H. M., Hessel, S., Hochscheid-Mauel, D., Aslanski, K., Deimann, M., & Kreuzberger, G. (2004). Kompendium E-Learning. Heidelberg: Springer. Oser, F. & Baeriswyl, F. (2001). Choreographies of teaching: Bridging instruction to learning. (pp. 1031-1065) In th V. Richardson (ed.). Handbook of research on teaching. 4 edition. Washington, D.C.: American Educational Research Association. Schnotz, W., Eckhardt, A., Molz, M., Niegemann, H. M., Hochscheid-Mauel, D. & Hessel, S. (2004). Deconstructing instructional design models: Toward an integrative conceptual framework for instructional design research. In H. Niegemann, R. Brünken & D. Leutner (Eds.), Instructional design for multimedia learning. Münster, New York: Waxmann. Schnotz, W., Eckhardt, A., Molz, M., Niegemann, H. M., Hochscheid-Mauel, D., & Hessel, S. (2004). Reconstructing instructional design models: Developing a heuristic advisory system as a tool for instructional design research. In H. Niegemann, R. Brünken & D. Leutner (Eds.), Instructional design for multimedia learning. Münster, New York: Waxmann. Rusman, E., van den Broek, A., Ronteltap, F. (2003). Guidelines for the evaluation of design patterns. Unpublished manuscript.