Johannesburg, 20 Aug 2018
While there will inevitably be some level of disconnect between what academia teaches students in analytics disciplines and what the companies that will recruit them after graduation require, one of the biggest challenges facing corporates taking on graduates is their lack of skill with the relevant analytics technologies.
This problem is created by the fact that the majority of tertiary education institutions teach their students data science capabilities using open source analytics software. This is only to be expected, explains Murray de Villiers, Senior Manager: Global Academic Programmes at SAS, since universities operate on a limited budget, and open source tools are seemingly free.
"Moreover, these solutions are quite effective in the classroom environment of a university. They work well for academic research performed only on small data sets, but they simply don't cut it in an environment where the data sets being analysed are huge and have a tangible impact on an organisation's bottom line," he says. "This is because these open source tools are not really scalable. A good example is the program known as 'R', which actually enables users to undertake sophisticated work, but unfortunately only with a small data set."
This creates additional challenges for graduates who are taught with these tools, he explains, but who then find that once they join the corporate workforce, their employer tends to use SAS solutions. De Villiers says the reason for this is because in the real world, decisions are make or break, so it is vital to use tools that are capable of handling large and complex datasets.
"Another challenge is that, sometimes, the universities teach students how to undertake spreadsheet analytics. While this is useful knowledge, it is once again something that is simply not scalable at all. Spreadsheets are clearly not built for the kind of large-scale data sets required to solve business problems. If you want to crack world-sized problems, you need analytical systems that are built to deal with these volumes of data," he adds.
"This is not to say that we think it is a bad thing for students to learn to utilise solutions like Python, R and even spreadsheets. It is always good to have knowledge of a range of tools, and more crucially, we want graduates to be aware of exactly what the limitations of these solutions are.
"At the same time, though, we feel it is important for graduates to also be exposed to industry-level solutions, the kind they will comes face-to-face with in the working world, as well. To this end, the SAS Global Academic Programmes is now providing local universities with a bundle of courses that can be integrated into their curricula, which can help students to learn their way around such tools."
The academic teaching materials and e-learning modules, explains De Villiers, are similar to those used for the training and education of SAS customers, with over 100 modules that have been developed by professionals in the US. SAS is now making this material available to universities free of charge, allowing them to integrate it into their courses as they see fit.
"What is notable here is that students who undertake these modules can, at the end of their degree, write the relevant SAS certification, and on passing, be instantly certified as a SAS data scientist. The course materials and e-learning run through both under-graduate and post-graduate degrees, so there is something there for everyone as well.
"Ultimately, though, the fact that students can complete their degree with a SAS qualification is something of a cherry on the top scenario, because they are able to add an internationally recognised certification to their degree, one that serves as the ultimate proof that they can handle themselves in real world situations, and demonstrates to potential employers that this is a graduate who has skills that are relevant for the industry," he concludes.
Share