Knowledge categories in e-learning: a meta-analysis
Type
presentation
Date Issued
2014-03-03
Author(s)
Zwingenberger, Anja
Abstract
Purpose
Learning effectiveness research increasingly turns to mixed modalities of instruction. We aim to provide guidance to designers of blended learning which combines traditional instructional methods with various multimedia formats.
Design, Methodology
We conducted a meta-analysis. In a categorical model of fixed effects we explored a matrix of three dimensions: 1. e-learning formats versus different non-e-learning comparators (e.g. print material or lectures), 2. deployment of e-learning in blended teaching (supplement) or as substitute for traditional instruction and 3. types of knowledge transferred. Particular attention is paid to potential confounders in trial design.
Findings
Analyses of 54 studies yielded a weighted effect size of 0.2. Ten variables correlate with effectiveness, of which four can be considered as indicators of quality and/or potential confounders and six as modulators. The former, including test methods, robustness of design, instrumentation and statistical power, help to assess original research. The latter, e.g. blended versus pure e-learning, comparator (learning aids of the control group in experimental situations), time on task, degree of interactivity and user control, help to design teaching concepts and media.
Practical implications
Variables providing guidance for rational choice of e-learning formats were assessed. In pure or blended e-learning, the category of knowledge transferred should guide formats, e.g., procedural knowledge is more effectively transferred by hypermedia or drill & practice than by tutorials.
Originality/value
This paper thus seeks to make a substantial contribution to the use, design and interpretation of e-learning.
Learning effectiveness research increasingly turns to mixed modalities of instruction. We aim to provide guidance to designers of blended learning which combines traditional instructional methods with various multimedia formats.
Design, Methodology
We conducted a meta-analysis. In a categorical model of fixed effects we explored a matrix of three dimensions: 1. e-learning formats versus different non-e-learning comparators (e.g. print material or lectures), 2. deployment of e-learning in blended teaching (supplement) or as substitute for traditional instruction and 3. types of knowledge transferred. Particular attention is paid to potential confounders in trial design.
Findings
Analyses of 54 studies yielded a weighted effect size of 0.2. Ten variables correlate with effectiveness, of which four can be considered as indicators of quality and/or potential confounders and six as modulators. The former, including test methods, robustness of design, instrumentation and statistical power, help to assess original research. The latter, e.g. blended versus pure e-learning, comparator (learning aids of the control group in experimental situations), time on task, degree of interactivity and user control, help to design teaching concepts and media.
Practical implications
Variables providing guidance for rational choice of e-learning formats were assessed. In pure or blended e-learning, the category of knowledge transferred should guide formats, e.g., procedural knowledge is more effectively transferred by hypermedia or drill & practice than by tutorials.
Originality/value
This paper thus seeks to make a substantial contribution to the use, design and interpretation of e-learning.
Language
English
Keywords
e-learning
meta-analysis
domains of knowledge
learning effectiveness
evaluation
HSG Classification
contribution to scientific community
HSG Profile Area
SHSS - Kulturen, Institutionen, Maerkte (KIM)
Refereed
No
Event Title
6th conference on e-learning Excellence in the Middle East
Event Location
Hamdan Bin Mohammend University in Dubai
Subject(s)
Division(s)
Eprints ID
227832