When I was deciding what to study for my undergraduate degree, the obvious “cool” choice was Computer Science. Everyone was talking about software engineering jobs, startups, and big tech salaries. But I chose Statistics instead — and it’s the best decision I ever made.
The Allure of Numbers Behind the Numbers
Computer science tells you how to build systems. Statistics tells you what the systems are actually saying. I was always more interested in the why than the how.
Growing up in Bangladesh, I saw data misinterpreted everywhere — from health statistics cited incorrectly in newspapers to election result analyses that fundamentally misunderstood sampling. I wanted to be someone who could look at numbers and understand what they truly mean.
Statistics is the Language of Research
Every field — medicine, economics, ecology, psychology — speaks statistics. When I started doing research, I realized that statistical literacy is the single most transferable skill a researcher can have. You can contribute meaningfully to almost any domain once you understand probability, inference, and modeling.
Machine Learning Needs Statistics
Here’s the irony: to do good ML work, you need a solid statistics foundation. Understanding the bias-variance tradeoff, knowing when regularization is appropriate, recognizing when your model is overfitting — these are fundamentally statistical questions dressed in engineering clothing.
My statistics training gave me the theoretical grounding that lets me understand ML methods at a deeper level than a purely engineering-focused approach would.
To Anyone Considering the Same Choice
If you love data, patterns, and making sense of the world through numbers — statistics might be the path for you. The job market is strong, the intellectual challenges are real, and the ability to contribute across disciplines is unmatched.
Choose the path that matches your curiosity, not the one with the loudest hype.