29-10-2012, 01:50 PM
Learning Objects Update: Review and Critical Approach to Content
Aggregation
IEEE.pdf (Size: 272.53 KB / Downloads: 40)
Abstract
The structure and composite nature of a learning object is still open to interpretation. Although several
theoretical studies advocate integrated approaches to the structure and aggregation level of learning objects, in
practice, many content specifications, such as SCORM, IMS Content Packaging, and course authoring tools, do
not explicitly state the aggregation level or granularity of learning content. The aim of this paper is to review,
compare, and amalgamate different content models for learning objects into a single and coherent learning
content hierarchy. To fulfil the objectives of this study, a substantive body of literature was reviewed and
analysed to identify current issues in the field of learning objects, and more specifically, learning object content
models.
Introduction
The structure and composite nature of a learning object is still open to interpretation (Metros, 2005; Knight, Gašević,
& Richards, 2005). Most of the definitions are shaped around general principles that govern the learning object
concept, such as reusability, learning intent, and context-independence. A typical example of such a definition is
provided by Polsani (2003), who defined a learning object as “an independent and self-standing unit of learning
content that is predisposed to reuse in multiple instructional contexts.”
Downes (2003) considers the size of a learning object to be important and provides debate about this. Cisco Systems,
one of the pioneers in the field of learning objects, tried to address this particular issue by suggesting that five to nine
information objects (collections of raw data such as text, video, images, and photos) can be combined to form a
learning object (Barron, 2002). Other authors approach the issue of the size of a learning object from an instructional,
time-based angle, suggesting that its size can be defined in terms of instructional time ranging from a 15-minute to a
two-hour learning experience (Downes, 2003; Mortimer, 2002). In contrast to the above approaches, however,
Currier and Campbell (2005) and Polsani (2003) argue that neither learning time, nor the physical size (in terms of
bits and bytes) is a valid criterion for determining the size and granularity of a learning object. They suggest that the
logical size rather than physical size is the appropriate concept for defining the size of a learning object.
Review of content models
Content models provide a framework for defining the structure, that is, the level of aggregation/granularity of
learning objects, and include the SCORM Reference Model (ADL, 2004), the aggregation levels defined by the IEEE
LTSC LOM standard (IEEE LTSC, 2002), the Cisco Systems RLO (Cisco Systems, 2003a,), the Learnativity content
model (Wagner, 2002), the content model defined by Schluep, Ravasio, and Sissel-Guttormsen Schar (2003) for their
prototype LCMS (learning content management system), and finally, the DNER & LO model (Currier & Campbell,
2005) (Figure 1). This section provides an overview of the different aggregation levels of learning content.
The SCORM content model
SCORM (sharable content object reference model) is a collection of related specifications and standards aimed at the
creation of interoperable, accessible, durable, affordable, and re-usable learning content (ADL, 2004). What is
particularly important about SCORM is the fact that it does not introduce new specifications or standards, but it coordinates
and refers to already established technical standards, specifications, and guidelines introduced by other
international organisations committed to e-learning standardization, such as the IMS, the IEEE LTSC, ARIADNE,
and AICC. This further justifies the wide acceptance and significant impact of SCORM on the e-learning industry
world wide. SCORM is defined by three separate but not mutually exclusive pieces of documentation: the content
aggregation model, runtime environment, and navigation and sequencing. From the three strands of the SCORM
documentation, the content aggregation model most relevant to this study as it defines the content components used
for creating a learning experience as well as the way these components can be aggregated into units of learning.
Conclusions
This paper presents a review and comparison of different theoretical accounts of the aggregation level of learning
objects from both an objectivist and a relativist perspective. It does not propose a new definition of a learning object.
On the contrary, it outlines the ambiguity with which the concept is used in the literature in terms of its aggregation
level and granularity. An objectivist and instructional design-based approach to learning objects can provide a more
concrete understanding of the contents of a learning object by both humans and machines (such as authoring tools
and e-learning systems). More work, however, needs to be undertaken to provide a standard specification of a
learning object and its aggregation level.