On March 8, 2022, Lancaster University hosted an online panel discussion on the career path of Women in Stats to celebrate International Womens' Day. The speakers were Anastasia Ushakova, Carolina Euan, Emma Eastoe, Kathryn Turnbull and Nicola Rennie.
My answers to some of the questions are below:
For me, it’s easier to answer this in terms of a typical week rather than typical day since I tend to have quite a varied schedule. In my job, I do a combination of data science consultancy and running training courses - mostly on programming in R. A typical week for me consists of two half days of training (online at the moment). I also spend sometime preparing for training, and developing or updating course material. In terms of consultancy, I spend two or three days a week on client work. That could be a wide range of things - it could be statistical advice i.e., helping clients decide which methods to use; it could be building dashboards; it could be reviewing code that a client has written. Then I also usually spend another day of my week doing internal consultancy, so projects that Jumping Rivers are working on internally. That might be building a website for a conference, running workshops at conferences, or automating some of our admin processes. That’s a typical week, but no two days are exactly the same and I can mostly pick-and-choose what I do on each day of the week.
During my undergraduate degree (in maths), I didn’t really do many statistics courses until my final year. But I liked it. It seemed more applied than most of the applied maths I’d spent the first few years doing. So I decided to go and do a stats PhD, which I did at STOR-i. I knew I wanted apply whatever I was learning so I was probably leaning more towards working in industry even before I started my PhD. During my PhD, I had an industrial partner - I collaborated with Deutsche Bahn, and I enjoyed that part of my PhD a lot more than the research side of things. I found myself wanting to spend time developing dashboards or playing with data more than I wanted to write papers (or my thesis). So I think at that point, I thought: okay, let’s find a job where I can spend my time doing the type of work I enjoy.
Most appealing? I really enjoy the variety of projects I get to work on. Both in terms of the application, and in terms of the actual technical work I’m doing.
In terms of hard (technical) skills, programming skills are the obvious one. You don’t need to be proficient in every programming language, but at the very least having an awareness of the programming languages that are commonly used (e.g., R, Python), and having done some work in one of them. I would also recommend getting used to working with version control early. It’s easier to practice using something like git when you’re the only person adding code to a project.
In terms of soft skills, I think being able to talk to people about maths without using maths-y words is really important. A lot of the time, clients will come for advice and they don’t necessarily know what the problem is. They also don’t usually have a statistics background. So being able to sit down and talk to them about what they want the end result to be, and what’s most important to them, is usually the first step in defining the data science problem that you’re actually going to be working on.
Don’t be put off for applying for a job because you don’t tick every box on the criteria. Data science is a quickly evolving field so even if you tick every box now, in six months those boxes will be different anyway. Most good employers aware of this and are willing to spend time developing employees. So just go for opportunities when they’re there.
- Posted on:
- March 8, 2022
- 4 minute read, 730 words