I shared this with my colleagues… the quote going around is that it is “easier to move a cemetery than to change a university curriculum”… LOL.

There are a lot of topics that the analytics theme embraces (think about all the analytics that goes into sports analytics… regression, simulation, optimizing!).  Among them, one that many  think is important is applied regression analysis.  I believe that we serve our students well if they know how to, when they graduate:

1) Run an MBA-relevant regression (using a software package… not by hand ;-).
2) Interpret MBA-relevant estimated coefficients (as partial effects, as dummy variables).  So many companies now do experiments, etc. where they are measuring the effect size, controlling for demographics.  Our students should be able to reasonably understand these things, regularly asking “have we controlled for…?”.
3) Leverage the evaluation of “statistical significance” to a process of asking questions.  They should be able to run regressions and notice, “hmm… not significant… does that make sense? Isn’t that interesting? I wonder why?”
4) Present regression analysis and interpretation with words and graphics, focusing on relevant information for the decision-maker.

I would like to see our analytics offerings shift in this direction.  It might not show up in a course count, but it would be a real change in direction, and it would be great if we agreed as a faculty on this direction, and could pool our talent to come up with a really great module/course on this.  Moreover, it would be fantastic if downstream courses could all integrate/reinforce this concept, since we are pretty sure that knowledge like this sticks poorly if not reinforced/repeated.

This is not to argue that decision-making software tools, simulation tools, linear-programming, etc are not also important analytical tools that our MBA students should have knowledge of.  My sense is that for the typical MBA graduate, a better understanding of regression analysis is now more useful than it was before, and our capabilities to teach it well are better than they were before (we have a raft of faculty across disciplines doing applied empirical analysis), and employers of our students are more likely to notice an improvement in quality if we focus more attention to this area.