Shareable Content Object Reference Model (SCORM) Research

While faculy at the Center for Statewide E-learning I was NSU's liason with the Academic Co-Lab of the Advanced Distributed Learning Initiaive (DoD) at UW Madison. Developing and defining the SCORM (Sharable Content Object Reference Model) was the pCo-Lab's rimary project at that time.

SCORM Diagram


The basic idea behind the Sharable Content Object Reference Model (SCORM) is to create a set of specifications that allow Content Objects (electronic learning objects) to be reused and to support searchability when archiving these objects. This is simple enough in theory but in practice a number of issues come up.


To solve the searchability problem the ADL (DoD Advanced Distributed Learning Initiative) developed a system of metadata tagging for the SCOs. This metadata can include such information as author, keywords, description, intended application, copyright and terms of usage, technical requirements (media "player" version, or image format, for example), addendums or alterations made (version and date), and even the various applications that this object has used for. The taxonomy for this metadata is a simple hierarchical structure containing some required elements and a large number of optional elements. A SCORM compliant SCO would have an XML manifest containing the metadata, and XML is searchable by any standard text based search engine making this document searchable from many different approaches.

Reusable/Context dependence

Reusability, however, is not so easily solved. Ellen Wagner, speaking to the ADL Academic Co-Lab gathering in Madison, Wisconsin, June 23, 2003, pointed out that reusability and context often behave as opposites on a single continuum. For a learning object or asset to be reusable it must be as flexible as possible. For a learning object to teach effectively it must provide context for its assets. For example, an image of the White House by itself is nearly context free and can serve in a wide variety of learning objectives. When I build a SCO that contains the image I add information, context. This might be historical information if the SCO is about American Civics, or engineering details if the SCO is about period architecture, but whatever the context I add I reduce the flexibility and therefore reuse of the asset within the SCO. Therefore, this tradeoff is fundamental to the concept of reusable learning objects.

SCORM Diagram

External References

Another issue this paradox brings up is that of external reference. We have become accustomed to hypertext in our research, the addition of links to information peripheral to the subject being directly addressed. We have also become somewhat accustomed to "broken" links and 404 "Not Found" messages. To link to a source outside of the SCO seriously reduces the reliable reusability of the SCO. Yet, as a course designer, I need to update materialoften, possibly even each time a course is taught. Further, I need to offer peripheral materials and additional resources for interested and motivated students. Somehow, I need to be able to allow students to access material not relevant to the SCO as it stands alone.

Sequencing and Navigation

SCORM Diagram

My project was based on reusability as a base concept and this presumption led me to certain conclusions. Immediately this meant breaking the course into subjects, concepts, and processes that could be taught independently of each other. The need for stand-alone content objects appears to be at odds with the need for coherent navigation through a course. This is compounded if the designer feels the need to address a multiplicity of learning styles in order to maximize student success. SCORM 1.3, in development then, was intended to address some of the sequencing issues. I chose to stipulate that for this whole course package some kind of organizing "envelope" would have to be built. (This envelope does, in a sense, become a learning object too, but is not a SCO because its use rests on the presumption of the use of all of the relevant Chapter/SCOs.) The envelope could contain introductory material, referential material, navigation and assessment material, and any further materials that might be relevant for the course in total, but not relevant to any of the single subject SCOs.

Using the paradigm of a textbook, the quintessential non-electronic learning object, the course subject can be broken into chapters, and most chapters can be broken into smaller sections. Each chapter would be a SCO, but each chapter-sized SCO could contain a number of smaller Sharable Content Objects. These smaller SCOs are commonly referred to as "aggregates" at this level.
Now each Chapter/SCO, in order to be reusable, must be able to stand entirely alone. Imagine, if you will, a series of boxes on a shelf, each one containing a Chapter/SCO. For the simplest navigation a user might work through these from right to left, (or number One, then number Two, then Three, etc.) Since one SCO can not call up the next (loss of independence, a different designer might use only one of the SCOs or want to use them in a different order) it is up to the end user to open one, finish it, and go to the next. This style of navigation can be referred to as a "page turner" or a "slide show." It is a simple sequential exploration. If you add pass-fail assessments at the end of each SCO you have "stair step" navigation, with competency at one level as the prerequisite for advancing to the next level. Unfortunately, this adequately addresses only one type of learning, directly linear. At the other end of the spectrum is the free-for-all approach. Imagine a textbook but without an assigned reading schedule. The user can jump about wherever they please. This chaotic learning approach does work with some users, but fewer than chaotic readers might believe. And there are a wide variety of learning styles in between. A course designer must be able to build any of a variety of navigational (and assessment) strategies into a course.

In practice much of the idea behind the SCORM has been washed out. To be compliant today little beyond an XML resource list is required. This list allows an LMS to ascertain that the required files are available, and that the format of the files is appropriate (hopefully), but little more. The Department of Defense, sponsor of the Academic Distributed Learning Initiative, sought to eliminate expensive redundancy in vendor supplied hardware and software. (One programmer from a major military instalation told me in 2003 that his base alone had 9 different CBT [computer based training] modules for training on the Beretta 9mm sidearm alone. None were remotely "compatible" with each other; all required specific and unique hardware, meaning a whole coputer dedicated to this piece of training, and expensive vendor support to run that particular training content.)

Return to Elearning page