The real change to the ADDIE process
Posted by Catherine Lombardozzi on September 12, 2008
I’ve written before that I don’t think the ADDIE model is dead, but I do believe it needs to be implemented more flexibly. The scope of that flexibility is going to be pushed to the edge as we move into different kinds of solutions. That is becoming more real to me based on a few recent events and ruminations…
At the E-Learning Guild Summer Seminar Series, Jeff Johannigman and Clark Quinn shared their experiences and thoughts on how to build games. I especially liked Jeff’s model which described the four steps of game creation as Analysis, Design, Implementation, and “Playtest and Balance.” According to Jeff and the game designers he consulted, the analysis phase is about defining the performance issues that are being targeted by the game, then exploring in detail the goals, possible actions, decision points, variables, resources, and possible situations that should be incorporated into game. You’re uncovering not only the goal of the learning you hope to achieve with the game, but details that will help you to design the game. At the design phase, you work out the setting, the interface, the player’s role, specific goals and subgoals, possible actions and consequences, and other details. Both phases require the active involvement of the artists and game developers – this isn’t something that a designer designs and then hands off for development. And then a miracle happens… The implementation phase includes all the development actions required to create a playable game – although not necessarily all programmed and pretty, mind you. The “playtest and balance” phase is when you test out all of your ideas and see how the game works. It involves your SMEs, target players, and malicious players (who can try to break it so you know what will happen). In this phase, your constantly iterating and tuning your game until it feels like it achieves the learning goals, appropriately challenging players with realistic decisions and realistic consequences, while still providing an element of fun.
That vision of game design concurs with one offered by Kurt Squire in a recent edition of Performance Improvement Quarterly. Dr. Squire did a qualitative study of the design of game-based learning and concluded that the process required providing a holistic model for clients (and setting their expectations), conducting iterative design, obtaining early user feedback, recognizing the increased importance of the artists and visual designers, and creating interdisciplinary design teams. Another echo regarding another way to think about design can be found in the Harvard Business Review article about managing creativity at Pixar, especially related to forming tight design teams jointly accountable for the outcome.
Michael Allen has already offered an iterative design and development model called “Successive Approximation” for e-learning products. And while I haven’t looked into it, I imagine the process of creating other web 2.0+ tools for learning are also quirky and iterative – but rapid as well.
I see the same thinking at work from our IT colleagues. Our organization has been implementing a similar iterative design and development model for IT projects – you may have heard about the Agile methodology that is strongly endorsed in the software development field. This methodology also lays out a plan for the entire stakeholder and technical team working together to define requirements, design, develop, and implement small pieces of a large project rather than launching a months-long process conducted entirely in the back rooms before anyone sees results.
Maybe you see where I’m going here. Adding that all up means…
As we continue to morph and flex the ADDIE model, we need to think about iterative design and development and creating teams of people to complete projects (rather than creating teams of people responsible for one function). It isn’t just about creatively implementing a design and development model to generate different kinds of solutions. It’s about a different kind of working relationship between and among designers, developers, artists, programmers, media specialists and – yes! – even clients – to pool creative energies and decision making. Now that’s turning ADDIE on its ear a bit. Those of you who are Jack-and-Jills-of-all-trades may not understand why that’s such a big deal, but those who work in organizations with functional specialties will see how revolutionary that is. That’s the real change we need to make in our overall process for creating learning. Now all I have to do is figure out what that really means. Mmm…
I wonder if you’re seeing some of these same changes… does your experience lead you to similar conclusions?
Sources:
Jeff Johannigamn and Clark Quinn (2008). Immersive Learning Simulations. E-Learning Guild Summer Seminar Series presentation notes.
Kurt D. Squire (2008). Video Game-Based Learning: An Emerging Paradigm for Instruction. Performance Improvement Quarterly 21(2).
Michael Allen (2006). Creating Successful E-Learning. Pfeiffer
cousinagam said
Thanks for this post, Catherine… I like Johannigman’s model, which I hadn’t seen before (or don’t remember seeing).
There’s a LOT of wisdom in the “playtest and balance” stage. I remember a very simple example years ago. Like all the other highly original training designers, I was building a Jeopardy-style game for reinforcing what had been learned the day before…
The first tryouts — our learners were salespeople — showed that their natural competitiveness could lead to one side just clobbering the other. That wasn’t the goal, so we tinkered with the game. You don’t HAVE to answer in the form of a question; you don’t HAVE to give the correct answer the chance to choose another… As you say, we tuned this simple framework to fit the learning goals.
The real heart of ADDIE is the heart of the systems model: the feedback loop. Too many people think “evaluation” means “give them tests.” Instead, I think it means track what happens, compare with goals, determine adjustments, and hop on for another round.
rani gill said
Hi Catherine – this is very thoughtful and informative post – thank you. I like the model, though I think there is a sub-section that is missing for me — designing with a specific technology in mind. When you design for a game, do the mechanics of the game matter? Is it a quiz game? a card game? a matching game? or some sort of systems systems game? or something else. This for me is where ADDIE falls down. The medium impacts the methods. Knowing what methods could work is a integral part of design. Maybe I’m being obvious here, but I never really see this addressed in instructional design models. Your insights would be much appreciated.
Catherine Lombardozzi said
Thanks for your question, Rani. I agree with you that design, technology, and methods are closely connected. Some colleagues have told me that we should design the experience and then figure out which technology can help us to deliver that experience – and there’s something to be said for that. But I think that we don’t have that luxury most of the time – we are limited by the technology that is available to us. We have to understand what is possible. So our design decisions and choice of techniques are greatly impacted by what is available to us, and at the same time if we get really creative, we can often get available technology to deliver something close. So yes, medium impacts methods, and methods impact medium.
But then the question is whether that fact impacts the ADDIE model. From my perspective, no matter what steps you actually take, you always do activities that can be characterized as assessment, design, development, implementation, and evaluation – you may not be entirely happy with the depth to which you go in each of these areas, but you do all of these activities. It isn’t possible to produce a product without them. That’s why I draw the ADDIE model in a sprirograph. So in that sense, methods and technology do not impact ADDIE at the high level – but methods and technology do impact exactly how that ADDIE process plays out and how long it takes.
rani gill said
Catherine – I’ve been reflecting on your comments. For me, the ADDIE model is too high-level to be really useful. It doesn’t really reflect my process.
If I were to create my own instructional design model, it would start with (1) identifying learning outcomes, (2) examine the nature of the knowledge, (3) identify the context and the messenger, (4) then choose a model based on the above. Then I would branch – and I use ADDIE for more procedural knowledge, and other design models for declarative, conceptual. I think of ADDIE should be seen as one tool within a larger process.
This thinking is a work in process. I will probably put something up in my blog, as a slog through the process of finishing my masters. Thanks for your perspective.
Is ADDIE Still a Relevant Model? « ADDIE-tood: Keeping it real in a virtual world said
[...] For example, Catherine Lombardozzi has given a lot of thought to changing the ADDIE model in her LearningJournal [...]