Describing the instructional design process in a single model is impossible. There are literally hundreds of models that try to capture the essence of instructional design. One wonders why we even need an ID model; why not just scrap models altogether and do what makes sense for a given project?
In truth, ID models help us identify what is generic and applicable across multiple types of projects. Models provide us with a common language to convey guiding principles for analyzing, producing, and revising learning solutions. They also help us plan resources, manage people, illustrate interactions between functions, and provide us with a picture of an ideal state. They are helpful, but even the best model has to be adapted to meet project requirements.
While ADDIE is generally represented as a simple five-step process, it is important to note that all instructional models are a form of ADDIE. After all, regardless of the type of solution being developed, there is always a need for some level of analysis, design, development, implementation, and evaluation. The difference between models is largely an outcome of where these steps occur in the process and how frequently they take place. For example, in a traditional waterfall model, each step only occurs once in a non-iterative manner—a true oversimplification of instructional design. When most people think about ADDIE, they are really thinking about a waterfall model.
In practice, there are two primary classifications that cover the most common instructional design models in use today: rectilinear and concurrent procedures models.
Rectilinear models represent instructional design as a series of boxes connected by straight lines, usually with one-way arrows—a true waterfall process. Some rectilinear models show a level of greater complexity, with the addition of one or more feedback lines. Unfortunately, these models don’t acknowledge the true complexities of ID, because they are rigid and cumbersome for real-life design challenges. In fact, most of these models emphasize a traditional, behaviorist approach to learning, which does not align well with the needs of most adult learners. The Dick and Carey Model is the best-known rectilinear model, and accordingly is often the model against which all other instructional design models are evaluated. The Instructional Development Institute (IDI) Model and Smith and Ragan’s Model are also examples of popular rectilinear models.
Concurrent Procedures Models
Concurrent procedures models show the instructional design process as sets of procedures occurring simultaneously, often with overlapping steps. Concurrent procedures models more accurately depict the practice of instructional design than do rectilinear models. However, they are sometimes difficult to explain to non-learning professionals, and they are not as easy to follow for project planning purposes. For example, the Morrison, Ross, and Kemp Model is a circular model that considers instruction from the perspective of the learner. The elements of the model are interdependent and allow for concurrency in the design and development process.
Most learning organizations use some type of rectilinear model. While they do not accurately show the richness and complexity of the instructional design process, they are usually easier to explain and implement. However, the inaccurate representation often causes cost and schedules to be underestimated.
Improving the ID Models
So where do we go from here? How do we improve our models to become more efficient?
More recently, some innovative models have gained a following that hold promise for capturing the complexity of instructional design while still allowing for effective communication about the process. These models place a greater emphasis on analysis and design, which are the most important processes of any development project. If the analysis and design is not well executed, then more cycles of revisions are required at later stages in the development process, when it is more costly and time-consuming to make the changes. Two such models are the spiral model and rapid prototyping models.
Originally conceived for software development, the spiral model combines both design and prototyping into each phase of the process. Each subsequent prototype increases in fidelity toward the end goal of the project. In essence, it recognizes that analysis, design, development, implementation, and evaluation is an ongoing process throughout the project lifecycle. The iterative nature of the model allows for stakeholder engagement and frequent end-user testing at each release, which helps mitigate the risk of the final solution not achieving the desired business outcome. The spiral model captures the richness of instructional design and provides enough detail for project planning. Barry Boehm’s Spiral Model of Software Development and Enhancement is the most famous spiral model and has been adapted by instructional design teams for development of complex learning solutions.
Rapid Prototyping Models
Rapid prototyping models are similar to the spiral model in that they recognize the value of producing a working prototype early in the design phase. However, there is variation between different models with respect to the number and frequency of prototypes that are developed. For example, some models focus on the development of a single rapid prototype to guide the development of the solution in an otherwise more traditional model. Other rapid models focus on developing multiple competing prototypes early in the design process. Then, based on feedback from stakeholders and end users, the best features from each prototype are integrated into the subsequent design for the solution. While rapid prototyping models are certainly an improvement over traditional models, many fail to convey the richness of instructional design, unless they capture all of the iterative cycles that are a necessary part of the process.
Instructional design and development is a dynamic, multifaceted process that is difficult to represent in any single model. While many organizations continue to rely on a simple rectilinear model to describe and plan projects, they do not fully account for cost and schedule risks that frustrate both stakeholders and development team members. Some newer models, like the spiral model, more accurately represent the iterative nature of design while considering the risk across the project. Regardless of the model that a learning team follows, the key is to evaluate the unique requirements of each project and adapt the model to fit the project.
Tim Youngman is a learning solutions manager with Caveo Learning. He spent 15 years as an instructional designer and eLearning developer, most recently serving as a custom learning solutions architect for Skillsoft. Based in the St. Louis area, he holds a master's in curriculum and instruction from the University of Houston.