Common Vocabulary for ID


Reigeluth, Charles M. & Alison Carr-Chellman. A Common Language and Knowledge Base for ID? Retrieved 2/6/2009 from


Reigeluth proposes that design theory differs from descriptive (learning) theory because it identifies methods versus cause-effect relationships; he also contends that design theory assists in the creation (versus mere description) of outcomes. His five sets of design theories parallels the ADDIE model.  He divides methods into several constructs (which he admits may not be all-inclusive):

  • Scope
  • Generality
  • Precision
  • Power
  • Consistency

These methodological constructs are contained within (and constrained by) the situational constructs of values and conditions (environment).


The argument that the cause-effect relationships of descriptive theory are probablistic is contradicted by Reigeluth’s earlier work that argues  that the methods of design theories, “are probablistic” (Riegeluth, Charles M. Instructional Design Theories and Models: A New Paradigm, Vol. 2 (1983). Mahwah, NJ: Erlbaum. p. 7); the earlier work seems more accurate. However, the idea to distinguish (and connect) design theory with learning theory is very useful. The act of situating methods in community (values) and environment (conditions) embraces constructivism and allows the methods to vary from behavioral to cognitive, depending upon the needs of the learner (developmental stage, as well as prior knowledge) and the content (including the goals).


ID/Project Management

Mc Daniel, K. and Liu, M. A Study of project management techniques for developing interactive multimedia programs: A practitioner’s perspective. Journal of Research on Computing in Education: 29.1 (1996). 29-48.

The mesh of ID and project management brought a refreshing practicality, and I especially appreciated the additions from Gentry on adoption (all of her supporting processes are essentially PM skills). Greet’s model didn’t add much; it seemed merely a merge of Dick & Carey with Gentry. However, the proposed 5 components were perfectly clear.

I was surprised not to see a discussion of Gantt charts and critical path in the Team Assembly and Management section. In the section on Evaluation, Marketing and Support, I have successfully used the conceptual differences between project management (the team) and project ownership (the client) to effect this critical transfer. If ongoing support is in the business model as a continuing revenue stream, it should be provided on a T&M basis.
The ID models seemed overly complex and somewhat idiosyncratic. However, the common desire for evaluation and revision is more easily accomplished with networked products (especially using agile and rapid prototyping) than with previous static delivery mechanisms.

As far as responsibilities, the following are key in my experience:

  • Keep big picture (outcome) in mind
  • Motivate team
  • Meet deadlines and budget
  • Free members from day to day
  • Break & sequence projects into manageable tasks
  • Use appropriate personnel (bring members along through smaller projects to develop independent decision-making)

However, my experience cautions that meetings should not necessarily “involve as many key players as needed.” Smaller teams may be more successful at problem-solving (see post on 2-player teams outperforming 4-player teams in online PBL). Also, while open communication channels are always essential, the project lifecycle I have found successful has 4 phases:

  1. Brainstorming – everyone talks, all ideas entertained
  2. Defining – everyone offers dependencies, tasks and times negotiated
  3. Monitoring – often, only key players present/talk
  4. Celebrating – offsite if possible, documentation review and lessons learned

Beginning the Development Process

The interesting aspect of Jenkins is that the website design process parallels ADDIE. The questions form the Analysis phase:

  • What are the audience segments (learners)?
  • What are the goals/uses (objectives)?
  • What are the metrics (outcomes)?

The Design phase involves creating prototypes and gathering feedback from users. The Development phase moves prototypes to full-scale production. Implementation is the site launch, and Evaluation is looking at the metrics.

The key issues are related to project management:

  • Establishing brand
  • Prioritizing bells and whistles
  • Architecting for future expansion
  • Obtaining approval/sign-off

Essential Instructional Design Elements

This chapter attempts to simplify the multi-step Dick and Carey model and succeeds because of its practical approach. The ID process is centered around 3 key ideas: design issues (elements); the design cycle (phases); and design collaboration (ACS).

Design issues focuses on 5 elements. Learners (at the center) are provided Outcomes, Activities, and Assessments (which are interrelated instead of being sequential) and wrapped by Evaluation (of the process).

The design cycle is viewed as 5 phases that spiral through the elements.  The phases are presented as an alliteration:

  • Define – identify learner needs – results in a project proposal
  • Design – determine assessment and scope/sequence – results in design document
  • Demonstrate – produce preliminary media – results in storyboards and a prototype
  • Develop – create instruction – results in a complete set of materials
  • Deliver – provide project

The spiral model is then aligned with a rapid prototyping model:

Create a vision – estimate the cost of each solution Define
Explore conceptual prototype – low-fidelity idea sketches Design
Experiment hands-on prototype – users “play” with mock-ups Demonstrate – mock-ups with limited function
Pilot working prototype – high-fidelity Demonstrate – “slice” prototype
Evolutionary development Develop & Deliver

The only issue that was unclear for me from the Rapid model is in the 3rd step: users are given hands-on mock-ups “to ‘play’ with but not to incorporate.” I wasn’t sure whether the users or the designers were not supposed to incorporate the prototype.

While we might wish to have at our disposal a formula to follow, the complex nature of learning requires the flexibility of a non-linear process. A superb quote summarizes the approach: these phases “are not distinguished by the type of task per se, but by the level of complexity associated with the tasks in that layer” (Tessmer).

Collaboration is explained through the ASC model: Ask/Assemble; Solve/Synthesize; Check/Confirm. The five phases each contain the ASC model although in earlier phases, different aspects of ASC collaboration are employed. For example (this was not in the chapter), Asking would be emphasized in every phase but especially in Define (discovering learner needs) and Demonstrate (uncovering the viability of solutions). Similarly, Solving would be used more in Design while Synthesis takes precedence in Develop.

Although ID is primarily associated with a team, the 5 elements are integrated even when an individual faculty member designs and delivers the instruction:

  • alignment among outcomes, activities and assessments
  • designed with learners in mind
  • continuous evaluation and revision

The final section asks a critical question ignored in most ID publications: when does a unit of instruction not warrant full-scale development? The answer rings true (the following are negatives of the situations presented in the chapter):

  • When the content is not stable
  • When the audience is not large enough to justify the costs
  • When communication among a team is not required
  • When it’s not important to test the instruction before it’s used

While the last situation (testing instruction before delivery) seems seldom applicable, the second and third situations are omnipresent in most higher education courses. The key take-away for me is the first situation: I have always thought that content should be separated into stable and dynamic types, and that stable content deserves learning object multimedia-rich development while dynamic content should be presented as “late-breaking news” in textual (and hyper-linked) format. This description now provides a more complete rationale for my argument.


I wasn’t sure if we were also supposed to post reactions to the chapters–Jason and Xavier did such a good job facilitating the discussion last night that it seems somewhat redundant. But I thought I’d mention a few things that stood out for me:

Dick and Carey split up the ADDIE model into 9 steps. Wiggins looks like it has 3 steps but actually offer more steps than Dick and Carey since Wiggins use sub-steps (or at least sub-questions). In that sense, I found Wiggins more useful. However (and this may be too cynical), all of the models seemed to be variations on the same ADDIE theme; in some cases, it seemed like the model creators were inventing a 6th or 7th or whatever step just to create a new model. That wasn’t true for all of them–Bates just as one example seemed to offer a new way of looking at things (because the model split course development from course delivery).

We talk about the 3 forms of interaction a lot with faculty when we help them plan their online courses. I suppose it’s possible we do so simply because Blackboard (or any of the mainstream LMS/CMS products are similarly organized around content, communication, and assessment); or perhaps it’s because Blackboard is organized around the three forms (which seems more likely given that Blackboard originally came out of Cornell). And so I was fascinated to see Dick and Carey mention presentation, participation, and assessment as key activities because these translate in my online world to lesson presentation (student-content interaction), discussion board participation (student-student interaction), and assignment/assessment evaluation (student-instructor interaction). Of course, I may be seeing patterns and similarities where none exist, or I may also NOT be seeing nuances that I should.

While I found more I liked in the Wiggins model than in Dick and Carey, that may be due to spending more time with Wiggins (if only because of the two chapters). What I particular enjoyed about Dick and Carey though was the idea of replication of results. This is critical to my job even though we deal with the (virtual) classroom. We spend a lot of time developing an online course, and if faculty leave, it may take two years to build a new course; in the meantime, another faculty member must step in and use the existing content to teach the class. In addition, as online courses grow in popularity, we find that the campuses are adding additional sections which are often taught by adjunct instructors because the sections are added at the last moment; in these cases, we want a consistency of content across all the sections.

Roles and Delivery
To me, Dick and Carey seems more geared to training than to education for two reasons. First, in many educational settings and increasingly in informal learning networks, the role of instructor and student shifts constantly. Using our class as an example, Jason was our teacher last night when he led the discussion. In an online game, sometimes my son is the team leader and sometimes he’s a follower (when he plays with his older brother). On the DEOS list-serv, I’m mostly a student but sometimes I contribute as a (self-professed) expert in some small area. Dick and Carey don’t take this learner-shifting into account (and in training, there’s no need to do so) . Second, Dick and Carey imply that design is independent of the delivery mode, an approach that if followed fails to take advantage of the unique characteristics of the medium and the delivery/learning environment.

Not Backwards at All
Wiggins makes a big deal of using a backward design approach but in actuality, it seems like all of the models (except the completely circular ones) do this: objective/assessment/content. However, the authors are 100% accurate that most faculty want to to start with the content. I’m surprised that no one used the maze analogy when talking about backwards design (solving a maze by working backwards). I thought the idea of uncoverage vs. coverage was clever, particularly when uncoverage was specified in the 2nd chapter as a process of dealing with misperceptions, grey areas, and core issues. I also liked the pairing of big ideas with core tasks (although I’m still pretty hazy on how to do this). And even though Wiggins seems to have a K-12 focus, I found much of the approach equally applicable to higher education. I disagreed with a couple of the analogies: designs are not like software–they are like movies or paintings (software is a tool); templates are not intelligent unless they adapt upon input.

The reason I really enjoyed the Wiggins readings (although I don’t think Dick and Carey would disagree with this) was the discussion of understanding versus knowing (although it seems difficult and complex to identify big ideas). The story of anatomy memorization results matching the curve for nonsense syllable memorization results was wonderful–I intend to use that at our faculty training next month. The tile analogy (patterns) was great, and I need to find another analogy for the transfer aspect of understanding. My own take: memorization is a process of dealing with facts and specifics which is equivalent to searching Google; understanding is a process of dealing with patterns and complexity which is equivalent to participating in social networks (not just Facebook–things like delicious and YouTube). This leads me to more questions (and no answers):

  1. Given the replicable power of systems models, can we devise a systems model that emphasizes learner participation (which might bring down the cost and timeframe–I guess this was the attraction of the Dorsey model for me)? Or will that approach dilute the power of the systems model (too many cooks)? This question might be a restatement of the Wisdom/Stupidity of Crowds issue.
  2. If designing courses for understanding recognizes that each individual brings his or her own socio-cultural background to the learning event, can we really design courses that work for every learner?

Who am I to disagree?

Even though Gustafson provided a useful factual description of ID models (I’d never seen one source that covered so many different models), I found myself making marginal notes of disagreement on almost every page (in fairness to the author, many of those quibbles were with the model creators, not the describer). Here are a few of my issues:

  • Crediting Boehm as a software developer is wrong—the article cited is about interactive video and multimedia development which is not what I’d consider software.
  • The three-part taxonomy of ID models does provide an organizational framework but one which seems limited and which is organized around the creator, not the learner or the outcome (note: look up Visscher’s taxonomy). For example, what about content built for individual online classes but designed for others to ruse (the concept of “master” courses or even the learning object approach)?
  • The classroom ID model emphasizes “selecting rather than developing.” Sort of. I see the actual practice as being one of deconstruction (which admittedly is selection) and creation of “missing” content (which is development) and then reconstruction (which is not development in terms of scope but IS in terms of sequence).
  • Gustafson describes one model as, “a general system view of development wherein all elements…may be performed independently.” (p. 29). I’m not certain systems theory says this; Dick and Carey’s systems model seems sequential albeit iterative.
  • Product ID models “must be usable (without) teachers.” Coming from the textbook industry in a previous lifetime, that statement doesn’t ring true.
  • Gustafson claims that interest in product ID models has increased with a growth in distance learning. The opposite holds true at the TeleCampus; we run more than 200 courses every semester and less than 5% use external products. Our online courses follow a classroom model, and my discussions with colleagues in online programs around the country suggest that we are very much like other universities in this regard (and in fact, many of us perceive it as an inherently limiting factor to scaling our programs).
  • Quality in the Nieveen model is defined (partially) in terms of validity. That sounds fine. But validity is equated with offering state of the art knowledge. What if knowledge is changing exponentially? Does that mean every course lacks quality?
  • Gentry’s two components are described as development and supporting, but I view the supporting components as actually practical constraints (budget, management).
  • I agree that the newfound interest in constructivism lies in technological advances. But Gustafson links those advances to the emergence of prototype design tools, when the advances are actually in online authoring and social interaction.

This sounds like I didn’t like the article at all—and that’s not true. What I found enlightening about Gustafson’s work (aside from the collection) was the emphasis on practice and the universality of the ADDIE acronym (even if no one will step forward to claim its creation).

As far as the models, I found several that I liked:

I liked Gerlach’s model because the activities are performed simultaneously but the initial specification of objectives and entering behaviors sets the context for creating those activities. In addition, the idea of a continuum of cues (from exposition to discovery) meshes well with my view of the importance of scaffolding (providing lots of help at the start and then withdrawing it). From an interactive view, the notion to vary student “group”: configurations based on strategy, space, time, and resources work not only in the physical but also virtual classroom and takes into account the importance of environmental factors which others models seem to downplay.

I liked Bates, but probably because it’s the basic process we’ve found that works at the TeleCampus: outline the course; create the modules; produce them; and manage the delivery in an online classroom. I thought DeHoog’s spiral approach was interesting, but it might be complex to implement (note: I wonder if this is related to Nokia?).

Despite my familiarity with Dick and Carey, Dorsey’s model was the most appealing because of the focus on rapid collaborative prototyping (although I also liked Diamond’s emphasis on a team approach pairing a faculty member with an instructional developer). To me, Dorsey’s model combines a systems approach with a practical, learner-centered thrust (get something in their hands) with an iterative and outcome-based orientation. This seems like the way Microsoft develops their products (I generally hate Microsoft but I must admit their products, at least in the Office suite, are really pretty good) or the way innovative Web (and especially Web 2.0) applications are developed: create something so you can get people using it and telling you what’s good and what’s bad, and then revise it over and over until it’s the best it can be (given your resource constraints). Infinite design timelines rarely produce infinitely better products. IMHO.