I’ve posted before about the Dick & Carey method of instructional design – while I was taking a class based on the Dick & Carey method. This method of instructional design is very popular because it represents a systems method of designing instruction. Click on the image below to see a diagram of how this method works:

But what does a systems method of designing instruction actually mean?

The definition of system is:

A group of interacting, interrelated, or interdependent elements forming a complex whole.

from thefreedictionary.com

What are these “interacting, interdependent elements” that may affect the development of instruction? Here’s a list from The Systematic Design of Instruction (Dick, Carey & Carey) along with my comments based on my experience in the world of designing technical instruction:

Individual Components of the System

• The instructor: What sort of training do they have? How much experience do they have with the product? How about the protocols, or the environment in which the product will be used?
• The learners: What sort of training do they have? What sort of information do they need about the product – will they be selling it? Will they be installing and configuring it? Will they be answering support calls from customers about it? Are they the customer? Or will the learners be a combination of all of these groups? Are they being forced to come to training, even if they think they don’t need it? Will they still have to answer customer calls and emails even if they are slated for training?
• Materials: What materials will be created for instruction? I develop training for software products that have some sort of revision every three months. Do we update our materials for each update of each product? What if a critical update is sent out for a product two weeks after we finish the materials?
• Instructional Activities: What instructional activities are needed? With software training, most of these activities are hands-on practice in labs built with the product being taught. But what should the activities be? How detailed should the lab instructions be?
• Delivery System: How should the instruction be delivered? Instructor led? Asynchronous eLearning? Synchronous eLearning? M-Learning?
• Learning Environment: In what kind of environment will the students be consuming the training?
• Performance Environment: In what kind of environment will the students be performing the activities that are taught during the training event?
• What have I forgotten?

Changing one component will impact the whole system

Each of these individual components work together with the other components. This means that if you change one thing midstream (lets say you make the decision to move from Instructor-led to eLearning, changing the delivery system), this will mostly impact other components of the overall system (the instructors, the learning environment, the materials, the learners all will most likely be impacted by the move from an Instructor-led to an eLearning delivery system).

And what happens if there is a component of the system that you haven’t even identified?

The systems way of thinking and performance

The systems way of thinking about instruction has been attributed to Larry Israelite (see Elliot Masie‘s Learning Rants, Raves, and Reflections 2004, review here). This way of designing instruction helps find performance problems so that the appropriate instruction can be designed. It provides a framework for systematically looking at a performance problem, and designing instruction so that the performance gap can be closed.

One reason it is important to apply a systems approach to instructional design is that one of the goals of instructional design is to  close human performance gaps. According to another one of my books, Mastering the Instructional Design Process (Rothwell & Kazanas) some of the things to consider when trying to lose those gaps are:

• Individual Performance: Does an individual worker have the right skills? Do they want to perform well? Do they have the tools to perform well? Do they have the ability to perform well?
• Work Group Performance: Can people work as a group? Is there a clear leader (that people are willing to follow?) Do individuals understand their roles? How do group members feel about the methods prescribed by the leaders to achieve group goals?
• Organizational Performance: Does the organization anticipate change? Does the org react well to change? Is there a culture of sharing in the organization? Is work being done in the most up-to-date fashion for the organization’s field?

If the real goal of “training” is to close performance gaps and enable a state of readiness in an organization, then it becomes pretty clear you have to think about a little bit more than creating power points, designing a lab, scheduling a classroom and sending invitations to students. A systems approach of designing instruction must be applied so that the effect on each individual component as well as factors affecting human performance are considered.

This semester I am taking a class called “Introduction to Instructional Design”. We are stepping through Dick, Carey & Carey’s textbook The Systematic Design of Instruction. The book examines the Dick and Carey model of instruction in detail. Here’s a diagram of that model (click the picture to see a larger version):

As with all models of instruction, the diagram shouldn’t be interpreted as a lock-step process. It is pretty fluid, you can go back and forth between steps as needed. The main points I take from the model is that you do constant analysis to ensure the instruction you develop is going to satisfy the instructional goal.

My degree program teaches instructional systems to people working in every sector: K-12 professionals, higher ed professionals, military folks and corporate professionals. Most of the corporate professionals develop training usually associated with HR, there aren’t many of us who do technical training. So while it’s good to have a map to ensure the effort and expense put develop training isn’t wasted, I have to wonder if this model is agile enough for a fast-moving technical education organization.

There are other instructional models that were developed specifically for technical instruction. Maybe I’ll write a post on that another day, it would be interesting to see if one has been developed to specifically fit in with Agile sprints and scrums.

We actually follow many of the steps in this model. One thing I (as a courseware developer) think we are missing is the learner and context evaluation. We use program managers to get that information to us from the managers of the students. Those managers deliver what they think the students should be able to do after consuming training, which is very important. It helps us set the correct performance objectives. But I have never been seen any learner or context analysis.

I can write courseware all day long, but if the students are starting from a different place than I thought they were, or if they can never get to training because they are too busy, does it matter how well the instruction is missing?

The biggest concern I have about the Dick & Carey model in a fast-moving technical area is that it takes alot of time to do the Analysis. The way we are learning this model, you have to be in the same room with learners to do the analysis and formative or summative evaluations. That would never fly with us, people at our company are just too busy. Not only that, but traveling all over the place do the analysis is too expensive. Besides, by the time we got back and compiled all the data, there would be a new version of software and we’d have to start all over again.

Why couldn’t I form a small community to pull all these steps together? The analysis would be there for all the stakeholders. The students could pop in and participate in guided discussions so that I have a real analysis of what it will take to get them to where their managers want them to be. We could pilot the materials in the community and get real-time feedback.

I think I know the drawbacks: people are still too busy to participate. People don’t trust online communities. Do we really need this extra step.

What do you all think? Does the Dick & Carey model move to slowly to be relevant for a fast-moving technical organization? Could it be modified and modeled in an online community? Do we really need to do so much analysis?