Becoming a Manager of Statisticians

John Johnson is the associate director of statistics at REGISTRAT- MAPI, a late-phase clinical research organization.

By now, we all know the demand for statistical skills is increasing rapidly, and the same is true for statistical leadership. Statisticians who are good at managing analysts and working with both project teams and clients are needed to make our processes better, work across disciplines within a company, plan projects with clients, and advocate for the needs of analyst teams.

Becoming a great manager starts at the beginning of a career, and the soft skills—competence and confidence in interpersonal interaction—required in management are necessary for managers and useful for technical-track analysts. In fact, analysts are well-positioned to pick up these skills by employing many of the same ideas used in computer simulation: Plan, Rehearse, Execute, and Reflect (but not too long).

For example, say you promised a project manager a delivery by tomorrow, but due to data problems, you will not be able to deliver until the day after tomorrow. Let’s see how these steps might look:

PLAN—write down on paper exactly what the situation is. Just get down all the facts, and perhaps even revise once for clarity. Remember to phrase everything in terms of actions and consequences. You also might consider adding a softener such as “I know I agreed to deliver tomorrow, but I just found out that …”

REHEARSE—work with a friend or mentor to do a dry run of presenting your case. Ask your rehearsal partner to come up with issues that may complicate your case. This kind of planning can not only help you strengthen your case, but also help you “think on your feet” during the interaction. You also might practice in front of a mirror, the same way you would rehearse a presentation.

EXECUTE—go to the project manager and present the case.

REFLECT—when the interaction is over, reflect on what went right and what went wrong. Focus specifically on your behavior during the interaction, rather than what was out of your control, and determine which behaviors you want to repeat.

Repeat this process for all sorts of interactions—working on a project team, contributing to department and company meetings, working with clients, and giving presentations. The final piece of the puzzle is to make a flexible plan of which technical and soft skills to pick up and a schedule for picking up these skills. An experienced coworker or supervisor can help with this in the beginning, as will paper and pen.

I created this plan by interviewing senior statisticians in the organizations I have worked for, and then took an afternoon to critically think about which skills I needed to pick up and prioritized them. For example, I placed a high priority on learning “the big picture” about how the statistics department fits into the company, which helped me identify which other departments had critical processes I needed to understand—data management, clinical strategy (which had the responsibility of writing reports), and so forth. In turn, I learned how to communicate with members of those departments effectively just by learning their language and rehearsing.

Revisit this plan every quarter or six months, because circumstances change. In executing the plan, volunteer for activities that take you just beyond your comfort zone. For example, my company needed to recruit several more statisticians, and our human resources group did not feel comfortable pre-screening the job qualifications. I volunteered to recruit statisticians for our group and had to become comfortable calling people out of the blue to ask them about their experiences. Not only did my phone skills improve, I ended up with the title of associate director within a year.

The world needs more statisticians who are willing and able to work well with our colleagues in both statistics and the sciences. With a little effort, you can pick up the skills necessary to be one of those leaders and perhaps lead a team of your own.