Leadership in Science
As scientists, we aren’t just researchers and educators; we are also leaders. Whether we are leading small teams, whole groups, or an entire experiment, it is worthwhile to consider what it means to be a leader. I’m sure we all would write down something different when asked “What does it mean to be a leader?”, but I’m also sure that we would all see the wisdom in what everyone else wrote. So, in that spirit, I’ve collected my thoughts about leadership in science.
So, what is a leader? I decided to go broad with my definition: a leader is someone who focuses the skills and efforts of a group of people toward a common goal. While this definition doesn’t explicitly give the power of decision-making to a leader, it certainly doesn’t preclude it. The standard “boss”, for example, falls easily into this definition. A boss leads by assigning tasks and keeping track of progress, using their power over their subordinates to get compliance. I’m sure there are people who would argue with me over this generalization, telling me that “good bosses (TM)” don’t wield power over their employees to enforce compliance. And I’ll admit that my choice of words certainly has a tad of a negative connotation. However, I’m not here to debate “good bosses” versus “bad bosses”; I’m just pointing out that a “boss” is a subtype of leader.
What I would rather focus on is another type of leader: a leader that doesn’t have power over their group. The kind of leader who can’t delegate tasks and demand compliance based on some sort of power dynamic. The kind of leader that other people follow by choice. We’ve all met people like this before. Maybe it was in school projects, or organizing community events, or within our social groups. These people lead by the virtues of their ideas and their social skills. These are the kinds of leaders that I want to look to while deciding what being a good leader means to me.
You’ll notice I changed the question. Before, the question was “What is a leader?”, but now, it’s “What is a good leader?”. Of course, you knew I was going to go there. It’s far more interesting to think about subjectively positive qualities than broad generalizations. The reason I want to look to people who lead without defined power dynamics is that, as scientists, we encounter far more groups without distinct power hierarchies than those with. Sure, you can think of the typical professor-grad student relationship as a case where there is an extreme power imbalance, but that’s hardly the predominant relationship in science. Far more often, we are interacting with peers and collaborators, over which we have no more sway than we can convince them to willingly give us. So, now that I’ve touched on why I want to answer this particular question, I should probably get around to answering it. What do I think makes a good leader?
First and foremost, even with explicit power, a good leader earns the respect and trust of the members of their team. Each member of the team should feel like the project is in good hands. The leadership should be seen as a positive influence on progress and results, not as something that needs to be overcome or dealt with (meetings that could have been emails, anyone?). Going a step further, team members should see their leader as a positive influence on their own work. If a project is like a puzzle, where each member is making individual pieces, perhaps a leader’s main job is to make sure the puzzle goes together. However, each piece should be no worse because of the leader’s involvement than if its creator were left alone.
Second, a leader should foster an environment where people can be productive, can learn, and can grow. This may seem obvious to some, and meaningless to others, but I think it’s important. People aren’t machines. We shouldn’t expect them to come onto a project ready to be maximally productive, requiring only maintenance to keep them going. People come onto a project and need to time to learn, to grow to fit the space they are filling. This expansion is why a person is better than a machine, by the way. A machine will never do more than it was built to do, but a person can adapt, fill voids, yield responsibilities to others, and generally become better at what they do as they do it. It is up to a leader to keep this in mind and support their team as each member grows into their roles.
Finally, a leader should create new leaders. What does this mean? A leader should create opportunities for other team members to assume leadership roles themselves. I would argue that being a good leader doesn’t just mean leading successful projects, but producing new good leaders as well. After all, in science, we don’t really finish with projects. As one winds down, answering whatever questions were answered, new projects spin off to answer new questions. A good leader will produce new leaders who can lead those new projects.
I didn’t expect to write this much when I sat down, but I’m happy with this. It’s important to really define what you’re striving for before you start considering how you will go about achieving it. But eventually, you need to make some plans. Here’s my thoughts on how I, personally, will go about achieving all those marks of being a good leader. This process is ongoing and needs to be adapted as I try new things, so consider this a perpetually work-in-progress:
1. I understand that, just because I’m in a leadership role, I don’t necessarily know more about each piece of the project than my team members. As such, I will take their advice and preferences seriously while trying to integrate them into the overarching course of the project.
2. I will maintain a light touch with individual work. This point is similar to the one above, but I broke it out because I think it bears special consideration. If someone is doing fine doing what they are doing, let them. Don’t fix what isn’t broken.
3. I understand that my personal goals aren’t necessarily the same as the project goals. I should, of course, try to accomplish both, but while the leadership hat is on, the project goals come first.
4. I will help people where I can, and I will encourage others to do the same. Science is about the free exchange of ideas and collaboration, and that should be reflected in my leadership. Additionally, by giving people the opportunity to work together, I am also providing opportunities for them to break in their own leadership hats.
I think this list is a good place to start. Like I said above, it’s important to revisit these questions, so they can be updated with new experiences and new thoughts. But, for now, I think this provides a solid framework for moving forward.
Suppose I put you in front of a light bulb and tell you I need to know what fraction of the time it is on. Easy enough, right? You can sit down and watch it, note the times it turns on and off, and the rest is math. But what if I were to put the light in another room and close the door. What would you do then?
This situation is one that I’ve spent almost a decade considering. How do we know what’s there without looking? You could, of course, open the door and watch directly. But a lot of the time, that’s what we would call a destructive measurement — the act of measuring the light bulb by opening the door spoils another measurement, perhaps the actual experiment that we are doing in that room. There’s nothing wrong with leaving the door open and watching the light, but we can’t do the experiment until it’s closed.
So let’s consider a possible solution, knowing just this. What if we just opened the door for one day out of every four? We could see how long it was on that day and keep a running tally. Then we could do some statistics and figure out a range like “the light is on for 4 hours, plus or minus 3, each day.” We only reduce the experiment’s runtime by 25%, and we can be relatively sure about how often the light is on.
“Ok,” you say, “that’s fine, but what about putting light detectors in the room and running them to a light outside the room you can watch instead?” Now we’re starting to get somewhere. We can’t look directly at the lightbulb, but what about some sort of proxy measurement? You can sit and watch the lights on the outside of the room, record their on/off times, and everything is great, right?
Well, not quite. Close, but not quite. How can you be sure that the lights outside the room are on at the same time the light inside the room is on? Let’s go back to the idea of opening the door once in a while. Except now, while the door is open, you’re watching both of the lights. After you’ve observed for a day and written down when both turned on and off, we’re going to do a bit more math and find out how often they were either both on or both off. This math is called the correlation between the two measurements. Maybe it’s 90%. Now for those next three days, as we watch the outside light, we can say, “there’s a 90% chance the inside light is in the same state as the outside light.”
Measuring the second light’s correlation with the first light provides us with additional data while the door is closed. These data will lead us to a better interpolation of what’s going on in the room. Interpolation is a mathematical process that aims to fill in what’s going on between two (or more) measured points. I’ll write another post about interpolation that gets a bit more into the nitty-gritty.
For context, what I have just described is very close to how we do the magnetic field measurement and analysis on the highly anticipated recently-published Fermilab Muon g-2 experiment. We have a very good magnetometer that can precisely measure the field (the trolley), but we can’t be storing muons while we are using it. So, what we do is use the trolley about once every four days. We also have a suite of other magnetometers that aren’t quite as good, but they can take measurements the whole time, even when storing muons. We use these other magnetometers, the fixed probes, to fill in the gaps (to interpolate) between the trolley runs and better determine the magnetic field.
Some Thoughts on Teaching
I believe that physics education aims not to drill students in equations and formulas that we deem necessary. Instead, it should be directed to teach the students how to learn, develop their own understanding, and share that understanding with their peers. To that end, my teaching philosophy centers around helping students develop physical models that they can use as a lens to see their future work through. I came to my conclusion about physics education through my experiences as a student, teacher, and, most importantly, a physicist.
My instructor for Introductory Mechanics my first year of college was incredible. The way he taught was like he was peeling back the curtain of introductory mechanics and showing us the beautifully simple mechanisms behind it. I was shocked to find out how often intro mechanics is taught as an example problem after example problems, a pedagogical machine after pedagogical machine. The way he taught the class wasn’t to show us examples of all the different exam questions we might see; he taught us how to approach a new problem that we hadn’t seen before. To this day, that lesson has given me so much confidence in my abilities to tackle new problems. I strive to instill that same level of confidence in my students by focusing class time on how to approach something new. For example, I pose difficult questions in class for students to discuss in small groups. Working in a small group in class alleviates pressure to answer the question alone while simultaneously exposing students to each other’s thought processes for approaching novel problems. Another way I instill confidence is to challenge students to write their own homework- and exam-style problems. I find that doing so is like asking them to pull back that curtain and gaze at the machinery behind it themselves; to see that all the problems they might see in their physics career are linked and that they need not fear the unseen, since it turns on the same premises they already know.
When I took multivariable calculus, the professor teaching the course would draw large numbers of axes directly onto the transparencies. He demonstrated that the topics we discussed applied to the standard 3D problems and any other number of dimensions. This concept stuck with me that the math we were learning wasn’t just applicable to the problems we did but could be extended to other cases. His cartoons of plots with 12 axes were models that showed me how the extensions worked. Today, I find that thinking in cartoons and simple models is one of the most important skills I have. When I’m teaching, I try to go beyond that, though. I believe that students need to be given the leeway to build their own mental models. This belief goes hand-in-hand with fostering confidence, as I mentioned above. Students gain confidence by seeing and understanding the underlying structures of physics and developing mental models. Similarly, well-developed mental models improve students’ confidence to keep attempting new problems and keep learning new things. A great way to build mental models is through interaction and experimentation with complex topics. I like to give students challenging problems that they can explore the nuances of and attack from many directions. Of course, it can be difficult to present purposefully challenging material without discouraging students or causing undue stress. That’s not my goal at all, and to prevent it, I fall back on transparent teaching policies and low-stress grading practices.
I don’t think that a teacher should withhold any part of a class from a student, with the possible exception of exam questions (but even then, I’m not convinced). One of the most significant parts of that belief is that grades should not be surprising to students. Between clear expectations for graded assignments and feedback from me, my students should know what grade they are on track to earn at any time and exactly what they need to do to change tracks if they want. To this end, I prefer the specifications grading method focusing on eventual mastery. In this method, students are only graded on whether they have demonstrated mastery over specified learning objectives. If they don’t pass any given assessment, they receive feedback about exactly why they didn’t pass and how to improve. Then, they are allowed to try again until they succeed. This grading style shifts the focus from “learn X, Y, and Z by given deadlines” to “learn X, Y, and Z by the end of the course, at a pace that works for you.” Students have the freedom to approach challenging problems without fear of how their grades might be affected.
I want to walk you through a day in my (in-person) intro mechanics class as a student. You come into class and grab the two handouts at the front. One is a blank sheet for scratching down notes to yourself, and the other has two big boxes on it with a dashed line for your name. I start the class by asking if anyone has anything to ask about a previous session or the required reading. After addressing any of those, I start going over the short assignment that precedes each class. These are just quick check-ins submitted online to gauge which topics students get and which they don’t. I go over any points I think were broadly missed, and then I pose a question to the class. Today, we’re covering the conservation of energy. I ask you to determine where on a simple rollercoaster with a loop the kinetic and potential energy are highest and lowest. I give everyone a minute to think and write anything down, then ask someone to volunteer their thoughts. Now I pose the next part of the problem — I invite you to determine the coaster’s velocity at several points. You get more time for this, a few minutes before I ask you to turn to your nearest neighbor and check to see what they think. You get a couple of minutes to compare your thoughts and talk to each other before I again call for someone to volunteer their thinking.
Now comes the fun part, I say. I ask you to determine the conditions that must be met for the loop-de-loop to be successful. You, your partner, and the nearest other pair get to talk about this. I remind you about centripetal force, then turn you loose to discuss together. I walk around, stopping to chat with each group about their thoughts and answering questions. After about ten minutes, I ask each person to summarize what their team discussed and decided on half their boxed sheet. After a few more minutes, I call the class back together, and we debrief about the problem. We probably get through another example like this before I start prepping you for the readings you need to do before the next class. Just before our time is up, I ask you all to think about why we used conservation of energy for this problem instead of conservation of momentum like we did before. You write your thoughts down on the other half of the sheet. As you leave, you drop off your filled sheet with me.
There are two points in the above scenario that I want to draw attention to: the use of scaffolded in-class work to prepare students for challenging problems and formative assessments to give students feedback in real-time. The scaffolding and check-ins start even before class time with assigned pre-class work. The students are built up from simple concept checks to basic examples before coming to class. Quizzes due before class give me a chance to check for their understanding and shore up misconceptions during class time before moving on to more in-depth examples. With a layered approach of practice, assessments, and feedback, the students can approach and tackle complex topics with confidence in a low-stakes environment. By the time that stakes are higher (they’re never very high because of my policy on mastery eventually instead of mastery right now), they have the confidence, skills, and internal model to demonstrate subject proficiency. More importantly, they will meet critical course goals and be set up for success in future studies.
Last updated on 3 April 2021.