Increasing the Analytical Skill of Business Students
In 2001, the proportion of business students at Australia's largest university who elected to study at least one quantitative subject in second year (first year statistics is compulsory) was 14%. Of those, only 1.3% continued this study into third year. Of the students who chose to go on to postgraduate studies, less than 1% elected to complete a quantitative-based course. By 2005, these figures had improved significantly to 45%, 13% and just over 5% respectively.
So whilst we can see the long-term trend is that the proportion of business students studying quantitative subjects is increasing, in the face of the huge explosion of analysis that is being made possible in the business sector through new technological developments, this still falls significantly short of meeting the resultant interpretational needs. What we have then is a substantial increase in the availability of output without a commensurate increase in the number of people who know how to interpret the value it contains.
Capacity Rising
I can remember in 1994 having to set a multiple regression task to run over night. It took 7 hours to run a single multiple regression model. Then, with technological innovation, software programs improved and regressions were able to be run in seconds. This meant that one could investigate hundreds of models to find the best one, which brought us back to spending hours on model development. Today, algorithms can be set up that automate the model selection procedure so that what previously took 3-4 hours now takes only seconds. The beauty of this is that it takes the ‘grunt work’ out of analysis and so leaves more time to interpret and generate insights. But where are the analysts who can do this? Where are the analysts who can correctly interpret the results in light of the management needs for which they were produced? Our business graduates? The contention is that whilst the levels of graduate numeracy have increased, our capability to produce analysis has increased faster. We are still left with a very large proportion of the graduate population who would have difficulty in interpreting or questioning statistical results.
Retraining?
In a newly released book, Competing on Analytics: The New Science of Winning, author Thomas Davenport, a long time proponent of the power of technology, addresses what CMOs should do if they are not skilled in analytical techniques: 'you better find some people who can explore the quantitative dimension of marketing for you.' Davenport also says that he thinks it would be wise to go back to school and learn quantitative approaches.
The challenge then remains as it has always been – how do we entice business students to study more analytical subjects in their degree? Perhaps the success of the highly analytical Jeff Bezos at Amazon.com might do the job!
