For the Love of God, Take Statistics: Course Recommendations for Undergraduates

What undergraduate classes will best prepare me for a career in science?
— Several summer students

This is not a question I ever asked. I selected my undergraduate courses in a blissful haze, overwhelmed by a list of Biology courses that stretched across pages and pages of the course catalogue. Did I really have to wait until senior year to take Integrative Organismal Biology? Sophomore, senior, same difference, right? Math requirements? What an irritating distraction from Behavioral Neuroscience and Immunology.

 One PhD later, and my regret over never taking a formal Physics course has mostly faded. I’m also far less irritated with my senioritis-driven decision to take Calculus II instead of any other math course in my final semester of college. I’ve learned the basics of physics required to comprehend the electrophysiology techniques I use almost every day, and I now know when to use a paired t-test versus a repeated-measures ANOVA. Would it have been easier if I’d taken undergraduate courses, instead of picking up this knowledge, piecemeal, from a combination of Wikipedia, user manuals, and methods sections? For the love of Ramon y Cajal, YES. 

Statistics: Not Just That One Math Class That Seems Easier Than Calculus

I asked a selection of Stanford neuroscientists what courses were most useful to their scientific careers, and which courses they most wish they had actually taken. For both these questions, mathematics was the most popular answer.

Kelly names statistics as the most useful class she took in undergrad, "because understanding which papers you should believe, and how to write credible and reproducible papers, are two of your top tasks as a scientist.”

Talia frames that same sentiment more succinctly, writing: “STATISTICS. For the love of God, take statistics. Because everything.”

Ian agrees, noting that linear algebra provided him with a framework for “just about every type of data or scientific problem.”

And one Stanford graduate student wishes she had taken more mathematics courses as an undergraduate, although since she’s “not strongly interested in learning theoretical math, it would have been great to take a class that applied statistics and mathematical concepts to biology.”

Our recommendation: Take a statistics course. Several scientists specifically recommended statistics courses offered through psychology departments (e.g. Stanford's Psych 252: Statistic Methods for Behavioral and Social Sciences), as these courses focus on practical applications. In the words of Allie, a recent Biology graduate: “Mathematicians teach you the theory behind each statistical test. Psychologists teach you which test to use and why using the wrong one will fuck your shit up.”

Can’t wait for an in-real-life class? Check out the online materials from Dr. Brian Caffo’s Methods in Biostatistics I, available free online courtesy of the Johns Hopkins Bloomberg School of Public Health.  

Computer Science: Disrupt your Biology Major

Not all computer science is focused on producing the next Google/Instagram/Uber, or disrupting the marketplace with data-driven, on-demand pineapple delivery services. Scientists use programming languages such as Matlab, Python, R, and Java to develop data analysis code. Learning Matlab (followed by Python and Java) has given me the flexibility to customize how I analyze my data; I’m far less constrained by pre-packaged (and potentially sub-par) software. David suggests learning coding ASAP: “I REALLY wish I had taken more computer science classes as an undergraduate. I’m still struggling to attain proficiency in computer science. Being able to employ computation to help you solve problems, whatever those problems may be, is invaluable.”

Beyond their utility as a research tool, day-to-day research, programming skills widen the career options for scientists at all levels. One Stanford graduate student credits the CS courses she took as an undergrad with kickstarting her scientific/graduate career. “Programming skills are the only true skill I took with me from undergrad. I didn’t have any lab bench work experience, just a lot of book-reading knowledge. Without programming, I don’t think I’d have eventually made it to the Stanford Neuro PhD program.” A postdoctoral colleague of mine credits his programming knowledge (Python, Matlab) with helping him land a Data Scientist position with a Silicon Valley startup.

So take CS courses. Newcomers to coding will benefit from general introductions to programming, for example, Stanford University’s Programming Methodology (CS106A). For online learning enthusiasts, CS106A lectures are available online via Youtube, iTunes, and Alex also recommends CS courses developed for scientists, if they are available. “The most useful course I took as an undergrad was an introductory CS course, in Python, that was focused on neuroscience/cognitive science applications. Basics of programming, in a fascinating and supportive environment, while learning a programming language of I have used countless times since? Check, check, check.”

Other Recommended Courses, Presented in Rough Order of Popularity

  • Physics
  • Engineering/Bioengineering
  • Psychology
  • Bioethics
  • Philosophy
  • Organic Chemistry

A Brief Note on Gender and Math/CS Education

When I attended Bryn Mawr, a women’s college, I deliberately avoided some of the gender politics that push young women away from the “hard” sciences (e.g. math, computer science, physics, and engineering). Several of the neuroscientists I talked to cited these forces as a factor in their decision to not enroll in the upper level mathematics and CS courses that they later wished they’d taken. One Stanford graduate student suggests that, “if upper level STEM courses have a bad reputation at your school, look elsewhere.” Extension schools (e.g. Stanford, Harvard, and Berkeley) offer online courses that can be quite affordable while offering a rigorous education. Coursera is another huge resource for structured self-directed learning.

Do You Have a Course Recommendation? Questions about Courses? Leave a comment below. 

1 Comment

Astra Bryant

Astra Bryant is a graduate of the Stanford Neuroscience PhD program in the labs of Drs. Eric Knudsen and John Huguenard. She used in vitro slice electrophysiology to study the cellular and synaptic mechanisms linking cholinergic signaling and gamma oscillations – two processes critical for the control of gaze and attention, which are disrupted in many psychiatric disorders. She is a senior editor and the webmaster of the NeuWrite West Neuroblog