Rob Santos

Greetings, fellow statisticians. I have advocated all year for us to ‘build community.’ The best way to do this is by supporting each other, especially during the pandemic. We are stronger and become more resilient when we act together. So, here is a story about how I learned to appreciate the value of helping others. But first, a little background.

I recently gave the keynote at StatFest 2021 and recounted to undergraduate and graduate students some of the ‘lessons learned’ I have gathered over a four-decade career. Preparing for that involved a bit of reflection, and it occurred to me I may not be your typical statistician, much less a typical ASA president.

In the autumn of my years, I will not be thinking thoughts like, “If I had only published an article on X in Y journal.” I’ll be reflecting on all the wonderful people I helped and wondering if I could have possibly helped others.

For instance, I am pretty sure I am the least-published ASA president in perhaps the last century. (I am guessing.) I am much more of a practitioner than a theoretician. My academic training (at the master’s level) was squarely in mathematical statistics. My experiential training has its roots in survey sampling and methods, and I’ve morphed into a policy researcher besides enjoying executive-level duties. Perhaps more importantly, I have focused my career on helping people by developing effective and efficient research designs and mentoring those who need coaching or someone to listen to them.

I decided early on that the secret to my own career advancement was making myself a source of value-added statistical input and/or professional advice. My professional satisfaction came from contributing to the success of other researchers and staff.

In 1982, I was hired to be the sampling statistician and manager of the sampling department at Temple University’s Institute for Survey Research (ISR) in Philadelphia. I was 28 years old, with a master’s degree in statistics and only a couple years of graduate student research assistant experience in the sampling section of the Survey Research Center at the University of Michigan. By taking this job, I was once again realizing my dream of “experiencing the real world.” Six years earlier, I had headed to Ann Arbor for graduate school without ever being north of Dallas, Texas. My move represented the first time I lived in the great Northeast. Suffice to say, I was a deer in headlights on multiple levels.

As for my new job, I was immediately responsible for drawing a large sample of new Social Security beneficiaries from databases provided by the Social Security Administration for a national area probability sample survey of 17,500 personal interviews, aptly called the New Beneficiaries Survey. I got through that project thanks to the marvelous sample design of Eugene Ericksen, who preceded me at Temple ISR.

Next, I had a really important and challenging sample to design and implement from scratch: the National Science Foundation Survey of Recent College Graduates. This study called for a mail survey of recent bachelor’s and master’s degree recipients in the US. The sample design called for randomly sampling 400 institutions nationwide and about 18,000 graduates. This involved a two-stage sample: After selecting the institutions, hard copy lists of recent graduates across a full calendar year were secured—often simply printed commencement lists from graduation ceremonies. The lists needed to be processed by hand, and systematic samples from numerous STEM fields by degree type had to be selected at different sampling rates to achieve the pre-specified representation for the study. And all this had to be done in weeks, not months or years. So, I needed lots of help. I naturally decided to use a cadre of undergraduate students to draw the sample.

During the process of screening and hiring a couple dozen students, it quickly became evident that my applicants often had little idea what they wanted to do for a career and were just looking for some part-time work to earn a few bucks. I could not fathom having them go through rote procedures to simply earn a few dollars. They deserved to get more out of the experience. So, I devised a plan to help them understand that work experience can be a source of personal and intellectual growth.

I developed protocols for list preparation, sample selection, and quality checks. I then trained my nascent employees. Having used my own creativity to establish a step-by-step sampling protocol, I was nervous about how effective it would be in practice. I had never done anything like it before. To address my reservations, I established weekly meetings of the group to debrief on their experiences with the workflow and sought their feedback on how to make the process better and of higher quality (i.e., reduce errors). The students eagerly embraced the challenge of making the process better and proved to be active participants in streamlining it while preserving scientific rigor. It helped solidify us as a team and instilled in them self-confidence. We were all in this together!

At the conclusion of each meeting, I asked the students to stop what they were doing and update their résumés to reflect their participation in these planning and operations meetings and the work they’d performed during the previous week. I noted they could now document their development of oral and written communication skills. They could also record their membership on a social science research project team. I further advised that they could state they engaged in process improvement activities and used creativity to address challenging technical and operational problems. I asked them to revise the written sampling protocols (which I would then review) so they could document technical writing experience. I had them switch off on supervision and quality checking so they could report that ‘higher level’ employee experience, as well.

The students were thrilled to be able to see how their part-time work gave them opportunities to develop skills and experience that would help them gain meaningful employment down the road, regardless of what they decided to do after graduation. Plus, it was all documented in their résumés in real time. I was delighted some students later decided to attend graduate school and went on to have amazing careers as business owners, social scientists, demographers, and programmers.

This early experience with students had a profound effect on how I approached my career. I found myself having no choice but to rely on student workers to fulfill my job duties and used that challenge to conduct rigorous scientific work while simultaneously helping students grow professionally. They were given opportunities to learn and gain experience in ways that transcended coursework. This represented a win-win I happened to stumble upon out of sheer necessity. The bonus for me is I found a way to pursue the two career passions I treasure most: statistics and helping people.

I have often used a strategy I call “growing leaders” over the last couple of decades. I partner with a junior staff person to author a blog or bid on a project where we are co-principal investigators. I provide technical/statistical expertise, and my junior colleague provides substantive expertise. Over the course of the project, I create a ‘space’ and provide the nurturing that allows my colleagues to take the lead in every sense of the word (budget, schedule, project team supervision, client contact, analysis, critical thinking, report writing). This is why you will seldom find a written report or blog listing me as the first author.

My professional reward is seeing those with whom I work blossom into leaders and scientists. I have found my niche, my way of fulfilling who I want to be as a professional. In the autumn of my years, I will not be thinking thoughts like, “If I had only published an article on X in Y journal.” I’ll be reflecting on all the wonderful people I helped and wondering if I could have possibly helped others. But that’s just me. I prefer a legacy of helping. What about you?