What is it about ADDIE that makes people so cranky?

In my preceding post I described how nice it was to have some F2F time with some of the bright and shining ID lights that I follow on Twitter. I also alluded to the fact that following last night's #Lrnchat,  I found myself getting a little cranky listening to others getting cranky and taking potshots at our venerable old instructional design process model, ADDIE, for...well, I'm not sure for what or why, exactly. 

To be clear - #Lrnchatters hate ADDIE. Among the #Lrnchat crowd, any time someone says ADDIE, everyone needs to take a drink. Usually of beer. Hard core. I know that part of the gleeful amusement about being in what Mark Oehlert calls the  "ADDIE-killing bizness, where bizness is good" comes in the spirit of knowing we need to get more, a lot more, from our learning designs. But still. Poor old ADDIE takes more than her share of hits.

It seems that the biggest objections to ADDIE are that it is linear. Not innovative. Doesn't branch. Worst of all...it isn't a learning model. In fact, at one point last night, one of the contributors dismissed ADDIE by saying she could use ADDIE to build a car. As if that was a bad thing.

Let's say it loudly and proudly - ADDIE isn't a learning model. Nothing about it says "learning model". It's a process model. It tells you what to do, in broad heuristic terms. It provides a systematic framework, a linear workflow - or, as Damon Regan suggested, a scientific method -  for solving problems related to designing and producing education and training experiences.

One presumes that if one is building solutions for education and training that learning is involved somewhere, somehow. Nevertheless it is ever so important to remember that instructional designers are not designing learning.They are designing learning solutions - resources, experiences, links, structured documents, multimedia and more, melded together into experiences that create the conditions needed for intentional, sustained changes in memory and performance to occur. Something tangible and real that can help people remember more, perform better, reach higher. In some respects, using a systematic process for building a learning solution isn't really all that different from a systematic process for building a car. A car gets us where we're going. So should our learning solutions.

I personally think that the biggest problem with ADDIE is that it is focused on process. Process is relatively intangible - hard to see, impossible to touch. And more often than not it is never-ending. It's hard to know when you are done. In the world of ADDIE, process is a  continuous circle of concern. Evaluation is just the next step on the path toward analyzing the next situation.

My suggestion is to think about ADDIE as a framework for producing what is known in software as a PRD, a product requirements definition document. A PRD describes exactly what you need to create your target product, in the same way that a blueprint give you directions for undertaking a construction project. 

Digital learning solutions ARE products; they are tangible and real and have requirements for being built, operated, maintained. They can be managed and shipped and stored and sold. They need to be attractive and compelling. They have a life-cycle that needs to be managed. They need to be marketed, promoted and sold. So if you are responsible for developing learning solutions perhaps you need to think more like a product manager - it is a lot easier to talk about value, investments, evaluation, life cycle management and the like when we are referring to something that is tangible and real. 

But before we move on - How about YOU tell ME - what is it about ADDIE that makes you cranky?

Comments

[Rob Pannoni]

You hit the nail on the head when you said ADDIE is like a software PRD. I've spent much of the last decade in the software industry. Haven't seen a real PRD in years. In my experience, hardly anyone uses them anymore. There just isn't time.

In fact, "software as a service" (SAAS) is making even the concept of a release obsolete. In the web 2.0 world, software changes go live weekly, if not daily. It's affectionately referred to as "perpetual beta." There's no way traditional learning development can keep up with that kind of pace.

Today, business runs on ad hoc, flexible, light, just-in-time and good enough. Yes, quality often takes a hit. But in the case of learning, we almost never measure learning outcomes anyway, so it's hard to show that it matters.

And, frankly, it may not matter. It's not clear the extra time/quality really buys better learning outcomes. It's simply the 80/20 rule. You can get 80% of the results with 20% of the effort.

The inability of traditional learning development processes to scale is why many of us are so excited about social networking, user-created content and other web 2.0 technologies applied to learning. It completely changes the equation.

[Kay Wood]

I wasn't one of the lrnchatters to discount ADDIE. Still ADDIE has frustrated me. As you stated ADDIE is a process. Barring any time constraints and the client being agreeable to an extended process, it works well. I haven’t found the agreeable client, yet.

I have worked as an ID in enterprise training and as a contractor for external clients. Internal clients were in a compressed time competitive market where training had to be created and turned on a dime. My external clients believe that analysis has been done on their part. That is why they want training, elearning, an e-seminar or some other learning event. They pay for everything but analysis. Time is the other critical constraint. Ah...the triple constraints of project management. Therefore development, design and implementation are what they pay for. We give them a level 1 evaluation for free. “Let’s take a short quiz to test your knowledge retention, OK?” Did you like the facilitator? Did you like the supporting materials?

A former college professor of mine naively told me once that I should refuse to work with a client who wouldn't let us approach learning development correctly with a solid up-front analysis. He was and still is an academic. Whereas I have never been other than the many years I spent accumulating degrees. There are probably organizations where ADDIE is standard. Sadly I have never worked at any of those organizations. I laughed at the professor. He never operated in the for-profit world.

On an intellectual level most know that it makes sense. It is a great practice. Just like I know that the Lexus is a great car. Maybe one of the best. I cannot afford the best. I, like my clients, can afford what I can afford. If I waited until I had the money for a Lexus, the day would never come.

As someone above wrote, maybe the linear, first, second, third, fourth, fifth thing is what gets on the anti-ADDIEs’ nerves. In a pressured environment, it isn’t like a flowchart, but more swim lane like with pieces of each step occurring concurrently at some point. Evaluations are occurring all throughout the lanes.

As a professional I do the best that I can with the constraints that I have. My goal is to always give the client what is best for their situation. There have been many times I didn’t, but most often it was because the client insisted on getting what they asked for which of course wasn’t what was best. By the time the RFQ is thrown over the fence for bid, the criteria is set. I don’t like it, but it is what it is.

I use ADDIE. My own process includes every letter of the acronym to some degree. ADDIE isn’t dead. It is like so many things today…it is not static.

[Don Bolen]

ADDIE is a model that turned into a jobs program for many ISDers. We must do it this way and that's why it takes so long. This days have passes.

[steve]

ADDIE isn't bad. BADDIE is bad:P It's the application that fails. ADDIE is the same dirivative intuitive distillate of every problem solving discipline in existence. Nothing in the framework, however, implies OR REQUIRES linearity. Folks with poor design and problem solving skills will quite likely not be able to synthesize process alterations to fit particular contexts and problem presentation. THIS IS THE PROBLEM.

I agree with John, it's about skills and specifically the lack of design skills that binds the industry to a largely mediocre existence. We see waves of attack against this model, that framework, or a set of ideas regularly. It's tough for us to lay the blame where it belongs.

However, I disagree with John S, at the implication that the lack of education in the field is at fault for skill deficiency. Fail design is fail, you can have a firm grasp on the rules and science and still be a poor designer. In fact, the worst designers I know have plenty of education and 'experience'. Design disciplines require affinities that most ID's just aren't packing.

Personnel selection PRIOR to that expensive education is key. Design isn't in the cards for everyone. Why don't we have a design aptitude evaluation for entry into any design vertical?

In addition to connecting the design dots, we've also got to get away from the cultural bent of ID's that seems to scream 'it's about me'. That's as anti-design as it gets and contributes to the piles of mediocre stuff we crank out every year, as an industry.

[Tom Werner]

Ellen, I think you make a good point that ADDIE represents process thinking...

...that a good way to think when confronted with a problem -- any problem -- is to analyze the situation, design and develop a solution, implement it, and then evaluate the results.

For me, however, calling ADDIE an "instructional systems design model" (as the Wikipedia entry does) is what grates on my nerves.

There's something that's a bit odd and self-important about it...

What if I were to invent the APSCCE Model (Analyze-Plan-Shop-Cook-Cleanup-Evaluate) and call it a "cooking systems design model"?

Or perhaps the APTRRE Model (Analyze-Plan-Takeapart-Repair-or-Replace-Evaluate) and call it a "household-repair systems design model"?

Are these really big intellectual contributions? Do they really deserve the stature of "models"?

If you didn't know how to cook, how much would my APSCCE Model help you?

If you weren't handy around the house, how much would my APTRRE Model help you?

Do I look smart -- or a bit silly -- if I go on and on about my APSCEE and APTRRE Models...?

[Howard]

Did not ADDIE first begin as a way to add an element of structure so that people did not just thoughtlessly teach the way they were taught. In that context it is a good thing, but like any structured response in a complex ecology, it's important to know when to modify, extend or totally leave the structure. I would think that assessment is an area of vulnerability in the process as a potential site of unexamined preconceived ideas that shape the assessment part and in turn effect the rest of the process. I bet that the real disagreements people have are in the root educational and assessment metaphors that guide and underly these preconceived ideas.

[John Schulz]

Ellen,

To start, I think you may be misinterpreting #lrnchater’s reaction to ADDIE. The frustration you perceive within many comments is due, I think, to the fact that many of us feel that the model is misapplied. I don’t hate ADDIE, but I can agree with Mark’s sentiments.

As you and other commenter’s have mentioned, ADDIE is a process model. It helps one outline a project – regardless of whatever the deliverables should be. And while ADDIE tends to be a waterfall model, there’s nothing that says you can only have one monolithic waterfall per project (though that’s how most of us have used it). Why not use a mini-ADDIE for each element of a project?

There’s also nothing that says you have to complete every step, in every phase, for every project. In fact, I suspect that the contents of each ADDIE phase look different for each of us. Use the framework as needed to fit within any type of project you may be working on.

The issue that I, and many others, tend to have is with those who proclaim to use the model but really don’t. Need your course tomorrow; well let’s just lop of this “A”. Not interested in continuous improvement; this “E” will have to go. Let’s minimize the time needed for that first “D” by making everything we do look exactly the same.

The rise of rapid development tools has caused our focus to shift from ADdiE to adDIe. Most organizations fail to remember that a solution is only as good as the level to which you define the problem. And if you don’t really understand the problem, design simply becomes an exercise in setting screen layout standards for PowerPoint.

Further complicating this, I think, is a genuine lack of Instructional Design skill. Informal surveys have suggested that more than 60% of us don’t have a formal education in ISD. Many of us came to the field ‘by appointment,’ as David Merrill calls it. As a result very few “IDs” truly understand the foundations of learning theory and cognitive science. We know learning theory through bullet points found in a presentation we saw at a conference somewhere.

The ripple effect is that if you don’t understand the roots of the theory, you don’t really know how or when to apply certain instructional methods. Analysis of the problem and design of a solution become oversimplified. And business results – truly impactful results of your learning solution – are difficult to prove.

Business leaders now have access to data we could only dream of a few years ago. And they are using this data to perform their own analysis on what activities are improving business results. As a result, more training functions are going to be asked to justify their existence.

So, it’s not that I hate ADDIE. But because others can’t justify its proper use they cut corners and turn us all in to order takers. Each time we misapply the model it becomes more difficult for us to push back on the business to justify their demand for ‘training.’ We get caught in a no win situation where we’re forced to build some product that won’t solve the root problem.

Bottom line – if you don’t know how to use the model, and you’re not willing to apply the WHOLE model (even in a rapid fashion), get out of the learning business so that the rest of us can develop meaningful solutions that at least have some chance of improving organizational capability.

(BTW – the drinking game isn’t about hating ADDIE either. It’s mostly a way to remind people to watch their use of buzzwords.)

Thanks for starting this conversation!

[Julie Dirksen]

This is something I've also noticed. A couple of thoughts:

1) It's the associations: ADDIE has an association with a very traditional type of instructional design — the old-school reduction of learning experiences to assembly-line, smallest-unit-of-behavior chunks. Conjures up images of very un-immersive, un-holistic, un-constructivist, deeply un-sexy sorts of instruction. You are absolutely right to identify it as a Process model, but it frequently doesn't get treated that way. There is nothing intrinsic about ADDIE that requires any particular type of design approach or learning philosophy, but I think ADDIE is adjacent to lot of those design approaches in Instructional Design 101 syllabi.

2) It's those pesky diagrams - I’m wondering if it’s the size of the boxes that gets people so riled. Even in the most imaginative of the graphical representations that try to get around the linearity problem, all of the categories are of mostly equal size. Is the implication that Design is only 1/5th of the process? That all the rest of that un-fun part of the process (I know there are people who genuinely enjoy evaluation, and god bless 'em) is 4 times as significant as the part that is the beating heart of the profession (we call ourselves Instructional Designers, not Instructional Analyzers or Instructional Implementers, after all).

[Dan R]

I think that the problems with ADDIE have well been documented in a far larger field than elearning - software development. It's a rather dull (as in blunt) tool for waterfall development and as such it fails on two counts - one is that it is far too long winded for many elearning projects - if your ID is your PM too then far too much time is sucked out on PM activities and the ID ends up having to take a back seat to reporting on progress and updating documentation. Moreover, it does not accommodate changes very well (unlike some agile techniques).

Of course, if you are working on very large projects then I can see that it would work better, though I suspect that more modern (and realistic) agile techniques may do so even better.

My own dislike of it is more to do with an employer that boldly proclaimed that they worked to the ADDIE model, but always proceeded with at the design step, taking the client's word for any analysis (often wrong) and then ending sharply at the implementation mode.

[Bill The Editor]

Some decision makers hate ADDIE because they hate the analysis step at the beginning. They reckon they already KNOW what's wrong. They take it personally when some nobody contradicts them. The IDs job is to go build them some training and be quick about it. The death phrase is to invoke "the paralysis of analysis." Some IDs (who should maybe know better) buy into this.

This is not to say that some people don't try to make a career out of the analysis phase. There comes a point at which analysis has to stop, and the rubber has to hit the road.Usually that's sooner rather than later.

There are other reasons people dislike ADDIE, but I think fear and loathing of Analysis is the biggest factor. People tend to just want to get on with it.