Universities Should Trade Specialization for Broad Approach
In my last post I described a variety of new analytically focused degree programs that are emerging at some universities. In this one I want to caution readers against the problem in universities that led us to the problematic analytical situation we are in today. It’s the issue of analytical specialization, and it’s endemic within most institutions of higher learning.
It was once said by Konrad Lorenz (http://en.wikipedia.org/wiki/Konrad_Lorenz ) that, “The specialist knows more and more about less and less and finally knows everything about nothing.” This description also applies to analytical specialization within universities. Faculty tend to specialize in particular analytical domains, e.g., optimization and linear programming, linear and nonlinear regression analysis, survival analysis, Bayesian statistics, categorical data analysis, etc., etc. ad infinitum. Once a particular set of analytical tools have been mastered, the faculty member rarely delves into a completely different analytical toolset.
Then the specialization of the faculty becomes the specialization of the student. If a student majors in, for example, marketing, the aspiring marketer tends to be exposed to analytical approaches that are useful within that function, such as cluster analysis, conjoint analysis, marketing mix analysis, pricing elasticity analysis, etc. He or she will be unlikely to be exposed to other analytical approaches used in supply chain management, finance, quality management, and so forth.
Such a student would then graduate and perhaps get a job within a marketing or market research group within a company. He or she would spend all of his or her time doing marketing-oriented analyses. It’s likely that there would be little contact with analysts in other domains.
What are the implications of this highly localized approach to analytics? One is that big, cross-functional problems—e.g., connecting supply and demand—don’t get addressed. Individual business functions optimize their own performance, but no one looks at how analytics can address the larger problems of the business. Analysts don’t get to expand their repertoires, and they solve the same problems over and over again.
If you’re working with an institution of higher learning on hiring analytical graduates, encourage the relevant faculty and administrators to give their students a broad analytical preparation. Ask them to create analysts who are not only deep in a particular analytical approach, but also aware of how different analytical tools are used throughout the enterprise. You’ll be able to solve tougher and more important analytical problems, and the graduates you hire will have more diverse and fulfilling careers.
