Numerous studies and statistics point to the fact that in just a few short years the need for people with analytics skills could significantly outpace supply.
With so much talk around the analytics skills gap and the growing market for analytic talent, we wanted to highlight a variety of avenues students and learners of all ages can explore to prepare for big data jobs. This blog series features interviews with professors, department leads and other educators who are seeding the market with analytical talent and directly impacting the talent management pipeline in this area.
Dr. Jennifer Priestly is Professor of Statistics and Data Science in the College of Science and Mathematics at Kennesaw State University.
How long have you been a SAS® user, and how were you first introduced to SAS? In an earlier chapter of my career, I worked for Accenture and with MasterCard. Both of those institutions utilized SAS for analytics. When I started my Ph.D. in 2000, I was required to learn BASE SAS programming to analyze data. I think I have used it almost every day since.
What SAS applications have you used in the past, and which applications are you currently using in the classroom? I tend to use BASE for all that I do – I really like the flexibility of writing my own code. I require all of my students to learn BASE SAS as well. When they finally see applications like Enterprise Guide – they come back to me and say “Are you kidding me? Why did you make me write all of this code…I could have just pointed and clicked!” I explain that if they know how to code in BASE SAS…then just about any SAS environment will make sense to them…they will be able to work with JMP, EG or EM…and they will understand how the software works. If we started with EG, it’s A LOT more difficult to learn how to code. The parallel I use is learning to drive on a stick shift…if you learn on a manual transmission…then you can drive anything. But if you start out on an automatic…a stick shift can be a bit of a mystery.
What does the analytics skills gap mean to you? We frequently hear about this analytical skills gap in the context of a large and growing demand for deep analytical skills combined with an insufficient “supply” of people who have these skills. I can give you a very real manifestation of this gap…At our university, we see health care companies, transportation companies, consulting firms and financial services companies all recruiting the same students. I don’t think this would have happened 10 years ago. So why now? I think we are seeing the employment equivalent of a “run on the bank”. Everyone is chasing the same talent at the same time…because they are all trying to solve the same problem…how to translate massive amounts of structured and unstructured data into meaningful information to improve decision making. And while some domain knowledge is helpful, the reality is that the core skills are the same – regardless of sector or industry.
But, I think there is another dimension to this issue…one that the academic community is falling over themselves trying to address. We have a fundamental misalignment of production and market demand. On the same day we can read an article about how new college graduates are saddled with student loans and can’t find a job…and then read an article about how there are millions of jobs that cannot be filled because there is insufficient talent. How can this be? Of the undergraduate degrees conferred in 2012-2013, the largest number went to “Business” (20%), Health Professions (10%), Social Science and History (10%), Psychology (6%), Biological/Biomedical Sciences (5%). So, what is obvious by absence? The core disciplines that would close the analytical skills gap! Mathematics and Statistics (<2%), Computer Science (<3%). But, data scientists are not Mathematicians…they are not Statisticians…they are not Computer Scientists…but they are some of all of these. I like the quote from Josh Wills (appropriately sent through Twitter) that the Data Scientist is the “Person who is better at statistics than any software engineer and better at software engineering than any statistician”. The “Priestley Corollary” to this quote is that the data scientist is the “Person who is better at explaining the business implications of the results than any scientist and better at science than any business school student” (can someone Tweet that out?). So, all of this would lead to “Data Science” becoming an interdisciplinary degree that would integrate Mathematics, Statistics, Computer Science, Business (or Health Sciences)…right? I think we all recognize that academia as an industry, is not doing a great job aligning our output with the needs of the market. The solution will include more interdisciplinary degrees like “Data Science”. This was the logic behind the development and launch of our Ph.D. in Data Science – which sits at the intersection of multiple departments and disciplines across the university.
I think we are starting to see universities entering into the lacuna of meaningful interdisciplinary degrees. Business Schools, Health Professional Programs and the Social Sciences are recognizing the need to integrate more mathematics and programming into their curriculum…Mathematics, Statistics and Computer Sciences need to integrate more communication skills, applied problem solving skills and visualization skills into their curriculum. And, there is no substitute for real world, applied projects, which can (and should) come from any (all) of these disciplines.
How do you think students benefit from learning about analytics and SAS? I tell my students that SAS is an “ante to play” in Data Science. There are a lot of different analytical platforms and languages out there…and they should learn several, but SAS HAS to be part of their analytical portfolio of tools. I believe this is true for two reasons – first, 95% of the Fortune 500s use SAS as their core analytical platform. While I can’t “guarantee” that they will get a job after graduation if they can program in Base SAS, I bet them a diet coke that it will happen (that is an important form of currency for me). The second is the latent learning that students acquire through SAS programming. I don’t like to use “point and click” options for students in the beginning of their studies. The problem with point and click alternatives is that they can generate too much output that they have no idea what it means. My opinion is that these are “cheap results” – they did not have to do much work to generate them, and so they don’t fully understand or appreciate what they mean. It is really hard to generate results “by accident” in Base SAS. And learning how to code, forces the students to translate the concept of what they are trying to accomplish into a methodical, linear series of code. It helps them to think in an organized, disciplined way…that point and click options don’t.
What’s the coolest or most impactful thing you've done using analytics? Perhaps an example you use in your class? At our university, we offer an undergraduate minor in Applied Statistics and Data Analysis – sort of Data Science-lite. The minor requires students to take a series of at least 5 3000 and 4000 level courses in applied statistics (most of these courses require Base SAS programming). So, this means that a student would major in Finance, and minor in Applied Statistics…or major in Chemistry…or Biology…or Psychology…or Communications…and minor in applied statistics. In any given semester, we have over 100 students who pursue statistics as a minor field of study across almost every college across campus. I teach a 4000 level course in Logistic Regression (using Base SAS). The course requires the students to extract, clean, load, transform and model a relatively large dataset (>1MM observations and >300 variables). I allow the students to work in teams…but I encourage them to organize into teams with people who are in a different major. As a result, I have Finance majors paired with Psychology majors…or Biology majors paired with Marketing majors. This is such an important – albeit latent – part of their learning in the class. They learn that (1) all disciplines have needs related to analytics – everyone needs to learn how to translate data into information and (2) people think differently about the same problem – Biology majors approach problem solving differently from Finance majors from Psychology majors. This is real life – we rarely work on homogeneous teams of people who think like we do.
What advice would you give students or adult learners interested in pursuing an analytics career? Learn how to program. And get LOTS of real world analytical experience through internships and projects.
Have you ever attended a SAS Users Group meeting or SAS Global Forum? If yes, please list them. I think I have attended every SAS Analytics Conference since 2010. I was the chair in 2012 and again in 2015. As a university, we participate in the SE SUG, and the Greater Atlanta SAS Users Group (which we have hosted a few times).
Please provide any additional information about yourself that you would like to share. Becoming an academic was a second career for me – I spent 11 years working in the private sector as a consultant for some great companies like Accenture, MasterCard and VISA EU in London. I “retired” and went back to school at the age of 33 to get my Ph.D. I have now been an academic since 2004. I can tell you that my 11 years as a working professional in the private sector make me a substantively more effective professor in the classroom – because I can integrate the theory and the application and draw from experiences that I had on client sites, working in teams, or unique challenges that you won’t find in the textbook. As we work to find ways to close this analytical talent gap, I would encourage anyone who has considered it, to bring their private sector experience into the classroom – I think industry/university partnerships are going to be an important part of the solution to solving the analytical talent gap problem.
SAS provides a wealth of resources for teaching and learning SAS. Check out the links below to learn more:
- Access free SAS software and resources for learning through the SAS Analytics U.
- Download SAS® University Edition software, and access 200+ free tutorial videos for learning SAS.
- Professors and educators can incorporate SAS into their academic offerings with resources from the SAS Global Academic Program, including curriculum consulting, certificate programs, instructor training programs, and student fellowship programs.
- Take your expertise to the next level with SAS Certification.
- See a list of institutions who offer master programs with a SAS focus.
- See a list of institutions who offer the Joint Certificate Program with SAS.
- We'd love to hear from you. Tell us about your experiences with SAS in the classroom. And check back soon for more upcoming interviews on this topic.