While taking V491, I realized I would need to use quantitative methods for my research. This was intimidating. I’ve never been one to willingly take math classes and didn’t know my way around a computer. I was steeling myself to pivot mid-semester if everything fell apart due to my lack of coding skills.
That didn’t happen. Despite not having a background in coding, I was able to learn R and feel comfortable using it in just a few months – and you can too.
Here’s my advice:
Youtube is your friend.
Whether you want to use Python, Stata, or R, YouTube is an invaluable resource. While it wasn’t easy at first, I took my time, slowed videos to 0.75x speed, and backtracked often. I was able to grasp the basics and even started writing some intro codes without references.
A word of caution: After my first Youtube video, I got overconfident, looked up “Poisson regression in R”, and promptly became overwhelmed and gave up for a week. I would strongly recommend taking the time to use some practice datasets before importing your data which will almost certainly need to be cleaned.
If you can, take a class.
I was lucky enough to take a class this semester in the political science department which gave me a solid background in R. It also helped me catch mistakes I would have otherwise missed. If you have the time and ability to fit a class into your schedule, I highly recommend it – the professor may also be able to act as a resource.
Use AI.
Seriously, use AI. ChatGPT is incredibly helpful with coding. If you describe your dataset and research, AI can often generate lines of code you can copy and paste into R. I wouldn’t recommend relying solely on AI, though – there were a number of times where I caught mistakes or reference errors thanks to the background I had in R by that point. However, I think AI can be a wonderful tool for beginner coders, and definitely provide some customized advice for your thesis.
Don’t be afraid to start over.
I reworked my models at least six times before I got the final version. It’s crucial to iron out the kinks, whether they’re related to data misalignment, coding errors, or simply forgetting to save your work (which happened to me once). The process can be frustrating, but it’s worth the effort to ensure everything is in its right place.
While using R involved a steep learning curve, I’m glad I took the leap. Building regression models allowed me to conduct the research I had envisioned. IUanyWare offers various statistical modeling tools, and if R isn’t your tool of choice, there are other options. If you’re considering quantitative analysis but feel intimidated, don’t worry—you can absolutely do it. Learning new data methods will open up new doors for your thesis research and beyond.
Darby Fitzsimmons is a senior at Indiana University’s O’Neill School of Public & Environmental Affairs.
Leave a Reply