The Real Problem with Confounding Variables

Maya Talukdar

[Interested in submitting? Check out the following link:]

If you think back to when you first learned about the scientific method, you probably remember a thing or two about confounding variables, which often lead to accidental association of two very different things. In fact, most scientists, even ones who are currently young students, probably have had it drilled into their head that correlation is not causation. And yet, I would posit that most of us have made this mistake when we think about the type of people science is meant for. After all, when you conjure up an image of an archetypal scientist, you probably imagine a lab coat and blue latex gloves, sterile apparel to match a sterile environment. Of course, science is a field based on logic and rationality, facts over feelings. And yet, it seems that this fundamental fact of what science requires has also been accidentally applied to what is required of scientists.

In our STEM classes, laboratories, and science clubs, it can be easy to feel as if we must leave other aspects of ourselves at the door. After all, C. elegans doesn’t care if you’re in a fight with your best friend or just failed a midterm; when it needs to be fed, it needs to be fed. But that doesn’t mean that those feelings aren’t important. While science requires rationality and logic, we too often assume that this means the best scientists are those who are able to power through everything they encounter in life while keeping a laser-focus on the lab. In turn, this can lead to serious doubt about one’s role in science when we face obstacles that follow us into or even occur because of our research; as stated in a recent commentary in Nature, “this is one reason junior researchers leave science — the demand to partition ourselves into separate entities can fragment our psyches.” On some level, the solution to this issue seems obvious; we all just need to accept that sometimes our personal lives can interfere with our science and not worry so much when it does.

I would argue, however, that being a good scientist not only means coming to terms with the fact that we are ultimately people first, with imperfect lives, aspirations, and history, but also actively incorporating these experiences into how we do science. Maybe you failed the first orgo exam while your classmates skated effortlessly by — now, you’ve learned the importance of tenacity, which will aid you when your first attempt at an experiment fails and you have to figure out what went wrong, make an action plan, and start again. Maybe you walked into a new research lab, looked around, and realized that no one else seems to look like you — now, you’ve learned the importance of advocating for yourself in a field that is not as diverse as it should be, as well as the importance of finding and being a good mentor with those you do connect with.

And, maybe, just maybe, you’ve walked into your dorm room after a long day of classes, problem sets, research, and volunteering and thought that you weren’t good enough to be a scientist, at Columbia or anywhere; I know I have. Great — now, you’ve learned the biggest truth of them all: much like the science we do, we can never be perfect. And yet, also like the aim of the science we do, we can still strive to attempt to discover something fundamental about the world, other humans, and ourselves, despite knowing that we will never find all the answers.

In the end, both science and our personal journeys aren’t about always making the right choices and achieving the optimal results, but rather recognizing that even our failures are intrinsically valuable for the mere fact that they bring us closer to discovering greater truths. As my PI told me this summer, a negative result may not be quite as desired as a positive one, but it still can add considerably to our knowledge. So, if you’re in the midst of midterm season, everything isn’t quite going to plan, and you’re wondering if this means you don’t belong, let me assure you: you do.