Thomson Delmar Learning  
      

Home   |    Contact Us   |    About Us    |    Press Releases    |    Careers   |    Site Map

 Browse Catalog

 Discover Our Resources

 Visit Subject-Specific Sites

The Next Generation of Corporate Learning

A Thomson Job Impact Study

"Blended Learning" is a popular concept in corporate learning, but lacks evidence in effectiveness -- does a blended learning model really increase employee performance?

This is the question addressed by a formal research study sponsored by Thomson and conducted cooperatively by the following corporations and academic institutions: Lockheed-Martin, NCR, Utah State University, University of Limerick, Ireland, Anoka-Ramsey Community College, Minnesota, Executive Service Corps of Chicago and KnowledgePool.



( Download is approximately 325K, and requires the Adobe Acrobat Reader)


The study sought to establish measurable differences in on-the-job performance achieved through a blended learning approach as compared to a traditional non-blended learning approach and to a ìno learningî environment.

In absence of a recognized standard for blended learning, a model was developed based on current best-practice learning research (Merrillís First Principles Concept). The result is a learning process that applies a ìlearn by doingî approach that progressively enables the learnerís independent performance. This structured learning approach is supported with multiple, flexible learning delivery methods, including online instruction, text-based instruction and reference, live and on-line mentoring and live instruction.

The study shows that a structured, well-planned blended-learning approach does dramatically improve employee performance on-the-job, both in terms of accuracy of task performance and in time to complete a task. Following are specific study results:

- Blended model learners were 30% more accurate in task performance than learners in traditional non-blended learning environments.

- Blended model learners performed tasks 41% faster than learners in traditional non-blended learning environments.

- Blended model learners were 159% more accurate in task performance than learners receiving no instruction.