However, I recently sent several on my team to attend Microsoft Windows Server 2003 MCSA and MCSE upgrade courses. Based on the feedback they gave me on these courses, I am changing my views. Sure, there was the usual theory content and some hands-on lab exercises, but what I really like is that the practical was also covered via investigative assignments and in-depth discussions on the results. For each topic area - say DNS or Active Directory - students were given a problem scenario that they needed to research in more depth. I really like this approach in that it mirrors the way that we have to work with the product in real life. Although I did my own Windows 2003 upgrade via self-study last year, I have a feeling that these classes could well have been valuable for me as well.

Interestingly, from what my folks have told me, not all participants enjoyed this new format. I guess some expected to sit there for five days and not have to think, and perhaps others didn't have the IT experience to be able to put these technologies in a business context (even though they were in an MCSA/MCSE upgrade class!). I suspect that not all students enjoy having to participate in class discussions as well. However, the more experienced students found this to be a very stimulating way to upgrade their skills.

The other side of the equation is how the class prepares for the corresponding certification exam, and here is where things get a little more muddled. In the distant past, there was a direct match between the course material and the exam content. Students could concentrate on the course as their only exam preparation and be safe in the knowledge that had everything there that they needed.

Since then, Microsoft's training classes have shifted gears, covering many topics but not exactly mapping what the exam covers. Presumably they expect students to read more widely and gain experience from working with the product in the field. The students at these Windows 2003 courses were told that these courses alone wouldn't fully prepare them for the upgrade exams, and that they would need to need to spend much more time in their preparations.

This does raise the question of what we expect from classroom education. Is it intended to improve on-the-job performance with the product? If so, this approach is fine. Or is it intended mainly as preparation for the exam? If this is the case, this approach is not so good. Of course, vendors do like to point to a correlation between certification and job performance, but as the products become more complex, there is a trade-off between the technical depth they can cover in the available class time versus an understanding of how these technologies can be used in practice. What is the best mix?

What do you think about the value of classroom education? What are your expectations of how well training courses should prepare you for exams? Let me know by posting your comments below.