Research Group Dynamics
December 2015 (perspective of an assistant professor)
A study at Google found five key dynamics that make for successful teams within the company: psychological safety, dependability, structure & clarity, meaning of work, and impact of work. These findings can be adapted to improving research group dynamics in academia.
In 2015, Google's HR group performed an extensive interview study that tried to answer the question, What makes a Google team effective? They found five key dynamics in effective teams:
I've worked at Google three separate times (two as a summer intern and one as a full-time software engineer), and I've also spent the past 1.5 years as an assistant professor mentoring 18 total students on research projects. Given my dual industry-academic experience, I'm well-positioned to brainstorm how to adapt these findings from Google into running an academic research group. The main caveat here is that I'm by no means an expert at leading such a group; I've been doing this for only 1.5 years, so I still have a ton to learn.
First off, I don't believe that these sorts of lessons from industry can transfer verbatim to academia since the goals of a product-driven company like Google are very different than those of a research-driven academic lab. However, with the right tweaks, I think we can get pretty close. Here's how I propose to adapt the five findings from the Google study to an academic lab:
1. Psychological safety: Can we take risks on this team without feeling insecure or embarrassed?
This is hugely important! I think it's absolutely necessary but challenging to foster this kind of psychologically safe lab environment in academia due to the status gap between professors and students. In industry, your direct manager is somewhat close in age or experience level to you, since they were probably in your role a few years ago; there are usually several layers of management between you and senior leaders. However, in academia, the gap between a senior professor and a junior student in their lab is huge. Also, students are young and relatively inexperienced, so their sense of self-efficacy isn't as developed as those of full-time industry employees.
I work really hard on providing a psychologically safe climate while still maintaining my authority as the group leader. The moment that my students are afraid of me and just (grudgingly) do what I say without question is the moment that creativity dies.
2. Dependability: Can we count on each other to do high quality work on time?
As academic researchers, we aim for creative breakthroughs that advance our field, but the day-to-day grind of producing research is a continual series of small-scale – often banal – tactical decisions. What should we try next? What's the next decision point? Should we try A or B? C or D? What's the next deliverable? Whom should we ask for feedback? When should we ask? How should we prepare for that meeting?
Thus, for a research group to work effectively week after week, month after month, everyone must deliver consistently and on time. This means that I make clear expectations for when we need to get which parts of the project done, and when my students need me to do something, I am clear about when I can get it done.
Flakiness is the absolute worst.
3. Structure & clarity: Are goals, roles, and execution plans on our team clear?
A research project may take months or even years to turn into a viable publication, often facing multiple rejections and pivots along the way to success. Thus, the day-to-day working life of a Ph.D. student can be notoriously vague and underspecified, which leads to uncertainty, stress, and existential angst. And undergrads are drawn toward more concrete, short-term endeavors such as classes, hackathons, and extracurriculars where there are clear and easy wins. How can we get students more engaged with academic research, given its intrinsically ambiguous nature? One of my main jobs as an advisor is to provide much-needed structure and clarity to counter the default of vague blahhness.
4. Meaning of work: Are we working on something that is personally important for each of us?
To me, this all comes down to finding the right student-project fit. Sometimes this means that the student finds something they are deeply passionate about. But often times – especially for undergrads or early-stage Ph.D. students – it means developing a project that can build the requisite technical skills and self-confidence to prepare them to tackle more substantive projects in the future, or to serve as a stepping stone to future jobs.
In other words, why is it a good use of my students' collective time to spend the prime years of their youth working on research rather than making far more money grinding in industry jobs? If I can't even answer that, then they have no reason to stay with me.
5. Impact of work: Do we fundamentally believe that the work we're doing matters?
It's crucial to convey to students where their particular project fits into the big-picture vision of our group, so that they know how their efforts contribute toward our shared goals. For instance, how does each project advance the state of knowledge in the given field? What is its relationship with related projects in the field? How may it affect the state of practice in the coming years? Why does any of this matter at all?
I'm not bold enough to claim that the research I do has any significant impact on the world (although a lot of people do use my software). If my students wanted to spend their time making a predictable and noticeable impact, they would be better off volunteering in the local community. But the purpose of attempting to move the needle on academic research is to seed the body of literature with novel, interesting, and potentially impactful ideas. Impact as a researcher should be assessed by very different standards than impact as a practitioner, so it's important for my students to set the proper expectations.