If you’re still in graduate school, there’s a lot you can do to help prepare for the possibility of a non-academic career down the line.
The most frequent piece of advice given by our respondents was to do an internship at some point in your graduate school career. This could be a summer-long experience or even a full year. For example, looking back, Maria D’Angelo shared: “I think an internship would have been helpful to get experience conducting applied research in a business setting.” You can use an internship to try out a new career path without needing to commit to it.
Taking classes outside of your discipline, especially in business, was another frequent recommendation. This is particularly important if you ever plan on starting your own company. Here’s what Ryan Dewey, who runs a consultancy business, shared he would’ve done differently: “I would have taken a business course. I would have learned about establishing a fee schedule, keeping books, what types of contracts I would need when working with clients. Learning how to be a better negotiator would have helped.”
These types of classes could also help you communicate with future colleagues. Katie Rotella noted: “I think taking a business or marketing class would have helped me to understand the mindset of the people I work with, as well as some of the lingo, but I don’t know that I would have learned more than the soft skill of how to speak their language.”
Graduate school is also a great time to pick up new technical skills, like programming and statistics, that you can later apply in different contexts. “I expected fairly early on that I would leave academia after my PhD, and so I focused a lot on acquiring skills that would generalize beyond academia, like statistics,” Brock Ferguson explained, “I’m glad I did this, but I wish I had done more. I would suggest that people publish one fewer paper per year and take an extra stats course, do online machine learning challenges, learn more programming languages, and build more fun things in graduate school while you have the time. Realistically, this will have a positive effect on your research anyway, so it’s not wasted time no matter which direction you go in.”
If you are thinking about pursuing a career in data science, which is among the most popular paths right now, make sure to pick up the skills you’ll need. Here’s what Ferguson advised: “If you are considering data science at all, you’re probably pretty good with statistics and R/Python relative to your colleagues in graduate school. This is a good start, but you aren’t just competing with other PhD graduates from your program. You’re competing against crazy smart software developers who have a deep tech background and, in the last few years, have picked up enough statistics and machine learning knowledge to get by. These software developers-turned-data scientists and engineers are very effective at what they do and serve a critical need for most companies who are just wading into data science for the first time. If you don’t have the software skills to stack up against them (i.e., you’ve never optimized a machine learning pipeline to handle millions of users’ data, you don’t know about how modern databases work, you only know one single programming language), you should take the time to get those skills before applying.”
For data science, here are some of the specific tools to learn as mentioned by our respondents: R, Python, JavaScript, big data storage and computation systems like Redshift, Hadoop, Hive, Presto, Spark, cloud infrastructures (mainly AWS services). In addition, one of our respondents, Nick Gaylord, recommended this recent primer to get oriented in the field.
If you enjoy working with quantitative data and are considering a future career that involves analyzing large and complicated data sets (even if this career is not data science), there are a lot of resources available to hone your skills if your university doesn’t offer relevant classes. For example, during her postdoc, D’Angelo “took a number of online courses in data science. Through these courses, I refined my programming skills in R and learned how to acquire and work with different data sources.” Likewise, Livins said she “used a lot of ‘learn to code’ type of sites – Learn Python the Hard Way and Mode’s interactive SQL site are both great resources. I approached stats by reading textbooks more than visiting websites. Gelman and Hill’s Data Analysis Using Regression and Multilevel/Hierarchical Models is a great one learning how to build some pretty sophisticated statistical models.”
A final piece of advice was to seek out mentors within academia who want to see you succeed no matter what path you take next. Here’s how Gaylord described his experience: “I owe a tremendous debt of gratitude to two of my professors/mentors in particular for setting me up to succeed in industry, and without whom I would have been super screwed. Jason Baldridge convinced me to stick with my NLP coursework and learn how to code (at least kinda), and persistently encouraged me to think about what I could do if academia didn’t work out. His was the only voice during my time in grad school to say this to me explicitly. Art Markman stands out as one of the best examples around of how social science expertise has real applicability to the business world. He’s offered me a lot of really valuable guidance about how to use my scientific skills to be successful in the business world. People like Art and Jason are hard to find in academia. Seek them out, and listen to what they have to say.”
During your graduate career, you’ll pick up valuable knowledge and skills that will be useful whether you continue on your current path or end up pursuing a non-academic career. Don’t hesitate to take the time and effort necessary to invest in yourself, and good luck, no matter what’s next!