Richard C. Zink is principal research statistician developer in the JMP Life Sciences division at SAS Institute. Prior to SAS, he spent eight years in the pharmaceutical industry. He is publications officer, 2018 chair-elect, and host of the statistics podcast for the ASA Biopharmaceutical Section. He is also author of Risk-Based Monitoring and Fraud Detection in Clinical Trials Using JMP and SAS and co-editor of Modern Approaches to Clinical Trials Using SAS: Classical, Adaptive, and Bayesian Methods.
I was recently asked to share my thoughts about SAS certifications—whether I thought they were valuable and, if so, which certifications were the most important. This is an important question, though I think it can be considered more broadly.
First, I have more than 20 years of SAS programming experience, but I do not have any SAS certifications. For full transparency, I do not hold professional certifications of any kind (though I can claim 6–8 scuba diving certifications). Even though I have been coding with SAS for so long, I still learn things all the time—new functions, tricks, and coding efficiencies.
Certifications, whether for SAS or any other area, are documented proof that you have some level of proficiency in a particular skill. Technically, you can view college degrees in much the same way—you have an organization willing to vouch for your skills after completing a set of requirements. Otherwise, potential employers or clients have to rely on letters of recommendation from former professors or employers, past clients who are willing to recommend your services, or actual code that illustrates how savvy a coder you are (such as a customized SAS macro with a number of bells and whistles). SAS maintains a website where any employer can visit to verify you have the skills you say you do.
Which certifications you try to earn ultimately depends on the industry you are trying to enter. Of course, if you are entering the pharmaceutical industry, the Clinical Trials certification will make the most sense. Here, part of the exercise is the language, itself, but also an understanding of data and analysis conventions within this particular industry.
But how do you prioritize certifications versus other skills or credentials? Of course, if a certification is required for a particular job, you had better obtain it. Otherwise, part of this will depend on the industry you enter and any perceived areas you think you are underdeveloped in.
For example, let’s say you have many years of R coding experience, but want to go into the pharmaceutical industry, where SAS is viewed favorably. Here, a SAS certification may be more important than SAS courses taken as part of a graduate program.
Or, suppose your initial career was in a nonstatistics field, but you go back to school to get a master’s degree in statistics or biostatistics to follow a more quantitative path. Here, an ASA GStat accreditation may be useful as a way to instill confidence in your skills. On the flip side, a PStat accreditation may be useful if you have been in the industry for a long time, as it documents you are making a concerted effort to maintain your skills and stay on the cutting edge of statistical science.
Certifications are one way to distinguish yourself from your peers. When entering the workforce, you are competing with other individuals to get a particular job. Most of us will get to check a master’s or doctorate degree box, so the question then becomes how else are you going to distinguish yourself? There are a number of ways:
- Certifications
- Internships in a particular industry
- Consulting projects with nonstatisticians
- Research or teaching assistantships
- Volunteering with the relevant sections of the ASA
- Experience in a less-common form of methodology
- Publishing
- Presentation experience
- Therapeutic experience
My opinion is that the single most important thing for young statisticians to do is to document in sufficient detail as much real data analysis experience as they have. This demonstrates the ability to apply what has been learned in the classroom, and, as many statisticians will tell you, the data is never as “nice” in real life as it is in the classroom.
I am also a big believer in developing a breadth of skills. For example, if you know one or two coding languages, you have documented that you have some capability with programming. Adding additional languages to your CV may not add extra value, since a lot of the differences in computer languages are in language syntax (unless there are specific job requirements for particular languages). Instead of developing a list of computer languages a mile long, highlight that you have taken a technical writing course, participate in Toastmasters or some other form of public speaking, or give presentations at scientific conferences. Or describe how you have volunteered with an ASA section to support a new initiative or had a leadership role in a student chapter. Or use that coding experience to publish a paper in the Journal of Statistical Software about a piece of software you developed. In short, communicate that you are well-rounded and have a diverse skill set.
Finally, don’t let the absence of a particular skill or certification—or lack of experience in a particular area—prevent you from applying for a specific job. Employers tend to list requirements that are so numerous and specific that the ideal candidate likely does not exist in real life. Describe the skills you do have. More importantly, highlight any skills or experience that make you unique and describe how they can benefit a potential employer. These other skills and experiences can often take a job or ongoing project in new and exciting directions—help the employer think outside the box!
Great article. As a grad. student, I asked Dr. Zink about SAS certifications as part of Asked Me Anything hosted by the Young Professionals Group of ASA! I am very excited he published an article elaborating his views!